a8cd0c01a6
Model: fpadovani/heb-hebr-10mb-after-ppt-shuff-dyck-10mb-ckpt500_seed3407 Source: Original Platform
4079 lines
110 KiB
JSON
4079 lines
110 KiB
JSON
{
|
|
"best_global_step": null,
|
|
"best_metric": null,
|
|
"best_model_checkpoint": null,
|
|
"epoch": 2.134543513080619,
|
|
"eval_steps": 500,
|
|
"global_step": 2000,
|
|
"is_hyper_param_search": false,
|
|
"is_local_process_zero": true,
|
|
"is_world_process_zero": true,
|
|
"log_history": [
|
|
{
|
|
"entropy": 4.68021821975708,
|
|
"epoch": 0.005339028296849973,
|
|
"grad_norm": 31.625,
|
|
"learning_rate": 2e-06,
|
|
"loss": 13.9711,
|
|
"mean_token_accuracy": 0.009074668399989605,
|
|
"num_tokens": 10658.0,
|
|
"step": 5
|
|
},
|
|
{
|
|
"entropy": 4.704092264175415,
|
|
"epoch": 0.010678056593699947,
|
|
"grad_norm": 32.75,
|
|
"learning_rate": 4.5e-06,
|
|
"loss": 13.9298,
|
|
"mean_token_accuracy": 0.00866024480201304,
|
|
"num_tokens": 21327.0,
|
|
"step": 10
|
|
},
|
|
{
|
|
"entropy": 4.7672216415405275,
|
|
"epoch": 0.01601708489054992,
|
|
"grad_norm": 36.5,
|
|
"learning_rate": 7e-06,
|
|
"loss": 13.7823,
|
|
"mean_token_accuracy": 0.007704685814678669,
|
|
"num_tokens": 32403.0,
|
|
"step": 15
|
|
},
|
|
{
|
|
"entropy": 4.958004379272461,
|
|
"epoch": 0.021356113187399894,
|
|
"grad_norm": 40.5,
|
|
"learning_rate": 9.5e-06,
|
|
"loss": 13.3459,
|
|
"mean_token_accuracy": 0.006649625021964312,
|
|
"num_tokens": 43471.0,
|
|
"step": 20
|
|
},
|
|
{
|
|
"entropy": 5.477138376235962,
|
|
"epoch": 0.026695141484249868,
|
|
"grad_norm": 38.75,
|
|
"learning_rate": 1.2e-05,
|
|
"loss": 12.4632,
|
|
"mean_token_accuracy": 0.0020860590098891406,
|
|
"num_tokens": 53724.0,
|
|
"step": 25
|
|
},
|
|
{
|
|
"entropy": 6.8245813846588135,
|
|
"epoch": 0.03203416978109984,
|
|
"grad_norm": 30.375,
|
|
"learning_rate": 1.4500000000000002e-05,
|
|
"loss": 11.6429,
|
|
"mean_token_accuracy": 0.0007934643130283803,
|
|
"num_tokens": 65115.0,
|
|
"step": 30
|
|
},
|
|
{
|
|
"entropy": 10.22093849182129,
|
|
"epoch": 0.03737319807794982,
|
|
"grad_norm": 5.46875,
|
|
"learning_rate": 1.7000000000000003e-05,
|
|
"loss": 10.8521,
|
|
"mean_token_accuracy": 0.0002344618726056069,
|
|
"num_tokens": 75902.0,
|
|
"step": 35
|
|
},
|
|
{
|
|
"entropy": 10.684853076934814,
|
|
"epoch": 0.04271222637479979,
|
|
"grad_norm": 4.0625,
|
|
"learning_rate": 1.95e-05,
|
|
"loss": 10.7564,
|
|
"mean_token_accuracy": 0.0011873395298607648,
|
|
"num_tokens": 87051.0,
|
|
"step": 40
|
|
},
|
|
{
|
|
"entropy": 10.69392204284668,
|
|
"epoch": 0.04805125467164976,
|
|
"grad_norm": 2.84375,
|
|
"learning_rate": 2.2e-05,
|
|
"loss": 10.5885,
|
|
"mean_token_accuracy": 0.006381610780954361,
|
|
"num_tokens": 98998.0,
|
|
"step": 45
|
|
},
|
|
{
|
|
"entropy": 10.600316715240478,
|
|
"epoch": 0.053390282968499736,
|
|
"grad_norm": 2.34375,
|
|
"learning_rate": 2.4500000000000003e-05,
|
|
"loss": 10.4445,
|
|
"mean_token_accuracy": 0.011505381856113672,
|
|
"num_tokens": 111207.0,
|
|
"step": 50
|
|
},
|
|
{
|
|
"entropy": 10.673304462432862,
|
|
"epoch": 0.05872931126534971,
|
|
"grad_norm": 2.0625,
|
|
"learning_rate": 2.7e-05,
|
|
"loss": 10.2638,
|
|
"mean_token_accuracy": 0.016619019117206334,
|
|
"num_tokens": 121039.0,
|
|
"step": 55
|
|
},
|
|
{
|
|
"entropy": 10.670036125183106,
|
|
"epoch": 0.06406833956219968,
|
|
"grad_norm": 2.03125,
|
|
"learning_rate": 2.95e-05,
|
|
"loss": 10.1533,
|
|
"mean_token_accuracy": 0.015064884908497334,
|
|
"num_tokens": 132604.0,
|
|
"step": 60
|
|
},
|
|
{
|
|
"entropy": 10.671291160583497,
|
|
"epoch": 0.06940736785904965,
|
|
"grad_norm": 1.7734375,
|
|
"learning_rate": 3.2e-05,
|
|
"loss": 10.0181,
|
|
"mean_token_accuracy": 0.03150726184248924,
|
|
"num_tokens": 142825.0,
|
|
"step": 65
|
|
},
|
|
{
|
|
"entropy": 10.688156223297119,
|
|
"epoch": 0.07474639615589963,
|
|
"grad_norm": 1.78125,
|
|
"learning_rate": 3.4500000000000005e-05,
|
|
"loss": 9.9346,
|
|
"mean_token_accuracy": 0.056240572780370715,
|
|
"num_tokens": 153121.0,
|
|
"step": 70
|
|
},
|
|
{
|
|
"entropy": 10.670577144622802,
|
|
"epoch": 0.0800854244527496,
|
|
"grad_norm": 1.7421875,
|
|
"learning_rate": 3.7e-05,
|
|
"loss": 9.8809,
|
|
"mean_token_accuracy": 0.05582823716104031,
|
|
"num_tokens": 164675.0,
|
|
"step": 75
|
|
},
|
|
{
|
|
"entropy": 10.664248943328857,
|
|
"epoch": 0.08542445274959957,
|
|
"grad_norm": 1.5703125,
|
|
"learning_rate": 3.95e-05,
|
|
"loss": 9.8352,
|
|
"mean_token_accuracy": 0.04962811656296253,
|
|
"num_tokens": 175146.0,
|
|
"step": 80
|
|
},
|
|
{
|
|
"entropy": 10.648977661132813,
|
|
"epoch": 0.09076348104644955,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 4.2000000000000004e-05,
|
|
"loss": 9.8124,
|
|
"mean_token_accuracy": 0.04548483826220036,
|
|
"num_tokens": 186177.0,
|
|
"step": 85
|
|
},
|
|
{
|
|
"entropy": 10.642164325714111,
|
|
"epoch": 0.09610250934329952,
|
|
"grad_norm": 1.6640625,
|
|
"learning_rate": 4.45e-05,
|
|
"loss": 9.685,
|
|
"mean_token_accuracy": 0.048678891733288765,
|
|
"num_tokens": 197143.0,
|
|
"step": 90
|
|
},
|
|
{
|
|
"entropy": 10.636518096923828,
|
|
"epoch": 0.10144153764014949,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 4.7000000000000004e-05,
|
|
"loss": 9.6182,
|
|
"mean_token_accuracy": 0.052110004052519796,
|
|
"num_tokens": 207979.0,
|
|
"step": 95
|
|
},
|
|
{
|
|
"entropy": 10.610146522521973,
|
|
"epoch": 0.10678056593699947,
|
|
"grad_norm": 1.640625,
|
|
"learning_rate": 4.9500000000000004e-05,
|
|
"loss": 9.5135,
|
|
"mean_token_accuracy": 0.0568073820322752,
|
|
"num_tokens": 218733.0,
|
|
"step": 100
|
|
},
|
|
{
|
|
"entropy": 10.569573020935058,
|
|
"epoch": 0.11211959423384944,
|
|
"grad_norm": 1.625,
|
|
"learning_rate": 5.2e-05,
|
|
"loss": 9.4144,
|
|
"mean_token_accuracy": 0.05693732798099518,
|
|
"num_tokens": 228887.0,
|
|
"step": 105
|
|
},
|
|
{
|
|
"entropy": 10.517056083679199,
|
|
"epoch": 0.11745862253069941,
|
|
"grad_norm": 1.5390625,
|
|
"learning_rate": 5.45e-05,
|
|
"loss": 9.3275,
|
|
"mean_token_accuracy": 0.05374570451676845,
|
|
"num_tokens": 240289.0,
|
|
"step": 110
|
|
},
|
|
{
|
|
"entropy": 10.480204105377197,
|
|
"epoch": 0.12279765082754938,
|
|
"grad_norm": 1.4609375,
|
|
"learning_rate": 5.7e-05,
|
|
"loss": 9.2538,
|
|
"mean_token_accuracy": 0.05631314069032669,
|
|
"num_tokens": 250548.0,
|
|
"step": 115
|
|
},
|
|
{
|
|
"entropy": 10.422281837463379,
|
|
"epoch": 0.12813667912439936,
|
|
"grad_norm": 1.4765625,
|
|
"learning_rate": 5.9499999999999996e-05,
|
|
"loss": 9.1616,
|
|
"mean_token_accuracy": 0.05645494796335697,
|
|
"num_tokens": 261669.0,
|
|
"step": 120
|
|
},
|
|
{
|
|
"entropy": 10.368197917938232,
|
|
"epoch": 0.13347570742124934,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 6.2e-05,
|
|
"loss": 9.0945,
|
|
"mean_token_accuracy": 0.05810702405869961,
|
|
"num_tokens": 272926.0,
|
|
"step": 125
|
|
},
|
|
{
|
|
"entropy": 10.291021633148194,
|
|
"epoch": 0.1388147357180993,
|
|
"grad_norm": 1.2734375,
|
|
"learning_rate": 6.450000000000001e-05,
|
|
"loss": 9.0261,
|
|
"mean_token_accuracy": 0.05110162906348705,
|
|
"num_tokens": 284173.0,
|
|
"step": 130
|
|
},
|
|
{
|
|
"entropy": 10.235691547393799,
|
|
"epoch": 0.14415376401494928,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 6.7e-05,
|
|
"loss": 8.9153,
|
|
"mean_token_accuracy": 0.05686397515237331,
|
|
"num_tokens": 294850.0,
|
|
"step": 135
|
|
},
|
|
{
|
|
"entropy": 10.190592193603516,
|
|
"epoch": 0.14949279231179927,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 6.950000000000001e-05,
|
|
"loss": 8.8582,
|
|
"mean_token_accuracy": 0.05346960201859474,
|
|
"num_tokens": 305727.0,
|
|
"step": 140
|
|
},
|
|
{
|
|
"entropy": 10.10048542022705,
|
|
"epoch": 0.15483182060864922,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 7.2e-05,
|
|
"loss": 8.7462,
|
|
"mean_token_accuracy": 0.058417992666363716,
|
|
"num_tokens": 317243.0,
|
|
"step": 145
|
|
},
|
|
{
|
|
"entropy": 9.970914077758788,
|
|
"epoch": 0.1601708489054992,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 7.45e-05,
|
|
"loss": 8.7387,
|
|
"mean_token_accuracy": 0.05338601246476173,
|
|
"num_tokens": 328334.0,
|
|
"step": 150
|
|
},
|
|
{
|
|
"entropy": 9.846445274353027,
|
|
"epoch": 0.16550987720234916,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 7.7e-05,
|
|
"loss": 8.6482,
|
|
"mean_token_accuracy": 0.054720787703990935,
|
|
"num_tokens": 339774.0,
|
|
"step": 155
|
|
},
|
|
{
|
|
"entropy": 9.73973569869995,
|
|
"epoch": 0.17084890549919915,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 7.950000000000001e-05,
|
|
"loss": 8.5543,
|
|
"mean_token_accuracy": 0.05573372282087803,
|
|
"num_tokens": 350211.0,
|
|
"step": 160
|
|
},
|
|
{
|
|
"entropy": 9.550614070892333,
|
|
"epoch": 0.1761879337960491,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 8.2e-05,
|
|
"loss": 8.4684,
|
|
"mean_token_accuracy": 0.054642071947455406,
|
|
"num_tokens": 360109.0,
|
|
"step": 165
|
|
},
|
|
{
|
|
"entropy": 9.380690956115723,
|
|
"epoch": 0.1815269620928991,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 8.450000000000001e-05,
|
|
"loss": 8.4131,
|
|
"mean_token_accuracy": 0.054827965795993805,
|
|
"num_tokens": 370911.0,
|
|
"step": 170
|
|
},
|
|
{
|
|
"entropy": 9.282039070129395,
|
|
"epoch": 0.18686599038974908,
|
|
"grad_norm": 0.76953125,
|
|
"learning_rate": 8.7e-05,
|
|
"loss": 8.3991,
|
|
"mean_token_accuracy": 0.060196897760033605,
|
|
"num_tokens": 381808.0,
|
|
"step": 175
|
|
},
|
|
{
|
|
"entropy": 9.13407211303711,
|
|
"epoch": 0.19220501868659903,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 8.95e-05,
|
|
"loss": 8.3091,
|
|
"mean_token_accuracy": 0.06106638610363006,
|
|
"num_tokens": 392175.0,
|
|
"step": 180
|
|
},
|
|
{
|
|
"entropy": 9.000042152404784,
|
|
"epoch": 0.19754404698344902,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 9.2e-05,
|
|
"loss": 8.3874,
|
|
"mean_token_accuracy": 0.049684269726276396,
|
|
"num_tokens": 403261.0,
|
|
"step": 185
|
|
},
|
|
{
|
|
"entropy": 8.931573295593262,
|
|
"epoch": 0.20288307528029897,
|
|
"grad_norm": 0.7265625,
|
|
"learning_rate": 9.45e-05,
|
|
"loss": 8.2853,
|
|
"mean_token_accuracy": 0.05462104417383671,
|
|
"num_tokens": 414992.0,
|
|
"step": 190
|
|
},
|
|
{
|
|
"entropy": 8.791546726226807,
|
|
"epoch": 0.20822210357714896,
|
|
"grad_norm": 0.6328125,
|
|
"learning_rate": 9.7e-05,
|
|
"loss": 8.2732,
|
|
"mean_token_accuracy": 0.05595530979335308,
|
|
"num_tokens": 425316.0,
|
|
"step": 195
|
|
},
|
|
{
|
|
"entropy": 8.729821968078614,
|
|
"epoch": 0.21356113187399894,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 9.95e-05,
|
|
"loss": 8.344,
|
|
"mean_token_accuracy": 0.05437092892825603,
|
|
"num_tokens": 437759.0,
|
|
"step": 200
|
|
},
|
|
{
|
|
"entropy": 8.712938117980958,
|
|
"epoch": 0.2189001601708489,
|
|
"grad_norm": 0.77734375,
|
|
"learning_rate": 0.000102,
|
|
"loss": 8.2844,
|
|
"mean_token_accuracy": 0.05534115061163902,
|
|
"num_tokens": 448066.0,
|
|
"step": 205
|
|
},
|
|
{
|
|
"entropy": 8.651909160614014,
|
|
"epoch": 0.22423918846769889,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00010449999999999999,
|
|
"loss": 8.2206,
|
|
"mean_token_accuracy": 0.055615832656621934,
|
|
"num_tokens": 458632.0,
|
|
"step": 210
|
|
},
|
|
{
|
|
"entropy": 8.63439655303955,
|
|
"epoch": 0.22957821676454884,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.000107,
|
|
"loss": 8.2606,
|
|
"mean_token_accuracy": 0.054173830151557925,
|
|
"num_tokens": 469224.0,
|
|
"step": 215
|
|
},
|
|
{
|
|
"entropy": 8.596088218688966,
|
|
"epoch": 0.23491724506139883,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0001095,
|
|
"loss": 8.2823,
|
|
"mean_token_accuracy": 0.05661116987466812,
|
|
"num_tokens": 480492.0,
|
|
"step": 220
|
|
},
|
|
{
|
|
"entropy": 8.625334930419921,
|
|
"epoch": 0.2402562733582488,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.000112,
|
|
"loss": 8.2398,
|
|
"mean_token_accuracy": 0.057639367505908014,
|
|
"num_tokens": 491325.0,
|
|
"step": 225
|
|
},
|
|
{
|
|
"entropy": 8.569988918304443,
|
|
"epoch": 0.24559530165509877,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0001145,
|
|
"loss": 8.2465,
|
|
"mean_token_accuracy": 0.0603976845741272,
|
|
"num_tokens": 502240.0,
|
|
"step": 230
|
|
},
|
|
{
|
|
"entropy": 8.592293834686279,
|
|
"epoch": 0.25093432995194875,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00011700000000000001,
|
|
"loss": 8.2035,
|
|
"mean_token_accuracy": 0.06263627484440804,
|
|
"num_tokens": 512710.0,
|
|
"step": 235
|
|
},
|
|
{
|
|
"entropy": 8.58193883895874,
|
|
"epoch": 0.2562733582487987,
|
|
"grad_norm": 0.71484375,
|
|
"learning_rate": 0.00011949999999999999,
|
|
"loss": 8.2513,
|
|
"mean_token_accuracy": 0.05789706408977509,
|
|
"num_tokens": 524023.0,
|
|
"step": 240
|
|
},
|
|
{
|
|
"entropy": 8.547989177703858,
|
|
"epoch": 0.26161238654564867,
|
|
"grad_norm": 3.234375,
|
|
"learning_rate": 0.000122,
|
|
"loss": 8.1422,
|
|
"mean_token_accuracy": 0.06462252773344516,
|
|
"num_tokens": 534612.0,
|
|
"step": 245
|
|
},
|
|
{
|
|
"entropy": 8.475037765502929,
|
|
"epoch": 0.2669514148424987,
|
|
"grad_norm": 0.68359375,
|
|
"learning_rate": 0.0001245,
|
|
"loss": 8.2036,
|
|
"mean_token_accuracy": 0.05881951674818993,
|
|
"num_tokens": 544866.0,
|
|
"step": 250
|
|
},
|
|
{
|
|
"entropy": 8.527704906463622,
|
|
"epoch": 0.27229044313934864,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.000127,
|
|
"loss": 8.2523,
|
|
"mean_token_accuracy": 0.055548783764243125,
|
|
"num_tokens": 556091.0,
|
|
"step": 255
|
|
},
|
|
{
|
|
"entropy": 8.524747467041015,
|
|
"epoch": 0.2776294714361986,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0001295,
|
|
"loss": 8.2536,
|
|
"mean_token_accuracy": 0.05479630120098591,
|
|
"num_tokens": 566946.0,
|
|
"step": 260
|
|
},
|
|
{
|
|
"entropy": 8.54754581451416,
|
|
"epoch": 0.2829684997330486,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.000132,
|
|
"loss": 8.1371,
|
|
"mean_token_accuracy": 0.060486416518688205,
|
|
"num_tokens": 577164.0,
|
|
"step": 265
|
|
},
|
|
{
|
|
"entropy": 8.525337314605713,
|
|
"epoch": 0.28830752802989856,
|
|
"grad_norm": 0.671875,
|
|
"learning_rate": 0.00013450000000000002,
|
|
"loss": 8.1768,
|
|
"mean_token_accuracy": 0.05711319632828236,
|
|
"num_tokens": 588744.0,
|
|
"step": 270
|
|
},
|
|
{
|
|
"entropy": 8.505339241027832,
|
|
"epoch": 0.2936465563267485,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00013700000000000002,
|
|
"loss": 8.1727,
|
|
"mean_token_accuracy": 0.055232659354805945,
|
|
"num_tokens": 600943.0,
|
|
"step": 275
|
|
},
|
|
{
|
|
"entropy": 8.492932891845703,
|
|
"epoch": 0.29898558462359853,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.0001395,
|
|
"loss": 8.1726,
|
|
"mean_token_accuracy": 0.06307518370449543,
|
|
"num_tokens": 611864.0,
|
|
"step": 280
|
|
},
|
|
{
|
|
"entropy": 8.451044940948487,
|
|
"epoch": 0.3043246129204485,
|
|
"grad_norm": 0.75,
|
|
"learning_rate": 0.00014199999999999998,
|
|
"loss": 8.152,
|
|
"mean_token_accuracy": 0.05931071788072586,
|
|
"num_tokens": 623314.0,
|
|
"step": 285
|
|
},
|
|
{
|
|
"entropy": 8.428306388854981,
|
|
"epoch": 0.30966364121729845,
|
|
"grad_norm": 0.69921875,
|
|
"learning_rate": 0.0001445,
|
|
"loss": 8.1763,
|
|
"mean_token_accuracy": 0.06025952883064747,
|
|
"num_tokens": 633509.0,
|
|
"step": 290
|
|
},
|
|
{
|
|
"entropy": 8.451824760437011,
|
|
"epoch": 0.3150026695141484,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000147,
|
|
"loss": 8.0983,
|
|
"mean_token_accuracy": 0.06582859642803669,
|
|
"num_tokens": 644655.0,
|
|
"step": 295
|
|
},
|
|
{
|
|
"entropy": 8.374590492248535,
|
|
"epoch": 0.3203416978109984,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.0001495,
|
|
"loss": 8.135,
|
|
"mean_token_accuracy": 0.05644083172082901,
|
|
"num_tokens": 655940.0,
|
|
"step": 300
|
|
},
|
|
{
|
|
"entropy": 8.408124351501465,
|
|
"epoch": 0.3256807261078484,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.000152,
|
|
"loss": 8.1809,
|
|
"mean_token_accuracy": 0.056555739417672155,
|
|
"num_tokens": 667074.0,
|
|
"step": 305
|
|
},
|
|
{
|
|
"entropy": 8.498903751373291,
|
|
"epoch": 0.33101975440469833,
|
|
"grad_norm": 0.703125,
|
|
"learning_rate": 0.00015450000000000001,
|
|
"loss": 8.2309,
|
|
"mean_token_accuracy": 0.06025502309203148,
|
|
"num_tokens": 677942.0,
|
|
"step": 310
|
|
},
|
|
{
|
|
"entropy": 8.455632591247559,
|
|
"epoch": 0.33635878270154834,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000157,
|
|
"loss": 8.1642,
|
|
"mean_token_accuracy": 0.06168182455003261,
|
|
"num_tokens": 689226.0,
|
|
"step": 315
|
|
},
|
|
{
|
|
"entropy": 8.403513431549072,
|
|
"epoch": 0.3416978109983983,
|
|
"grad_norm": 0.76171875,
|
|
"learning_rate": 0.0001595,
|
|
"loss": 8.1786,
|
|
"mean_token_accuracy": 0.059801208600401876,
|
|
"num_tokens": 700830.0,
|
|
"step": 320
|
|
},
|
|
{
|
|
"entropy": 8.456379795074463,
|
|
"epoch": 0.34703683929524826,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000162,
|
|
"loss": 8.1377,
|
|
"mean_token_accuracy": 0.06001129299402237,
|
|
"num_tokens": 712677.0,
|
|
"step": 325
|
|
},
|
|
{
|
|
"entropy": 8.363616943359375,
|
|
"epoch": 0.3523758675920982,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.00016450000000000001,
|
|
"loss": 8.2395,
|
|
"mean_token_accuracy": 0.05579938478767872,
|
|
"num_tokens": 723833.0,
|
|
"step": 330
|
|
},
|
|
{
|
|
"entropy": 8.497436428070069,
|
|
"epoch": 0.3577148958889482,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00016700000000000002,
|
|
"loss": 8.1771,
|
|
"mean_token_accuracy": 0.059333324804902075,
|
|
"num_tokens": 735526.0,
|
|
"step": 335
|
|
},
|
|
{
|
|
"entropy": 8.438710403442382,
|
|
"epoch": 0.3630539241857982,
|
|
"grad_norm": 0.69140625,
|
|
"learning_rate": 0.00016950000000000003,
|
|
"loss": 8.1223,
|
|
"mean_token_accuracy": 0.06136825978755951,
|
|
"num_tokens": 745934.0,
|
|
"step": 340
|
|
},
|
|
{
|
|
"entropy": 8.391829586029052,
|
|
"epoch": 0.36839295248264814,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.00017199999999999998,
|
|
"loss": 8.0837,
|
|
"mean_token_accuracy": 0.06251248680055141,
|
|
"num_tokens": 756693.0,
|
|
"step": 345
|
|
},
|
|
{
|
|
"entropy": 8.393558979034424,
|
|
"epoch": 0.37373198077949815,
|
|
"grad_norm": 0.74609375,
|
|
"learning_rate": 0.00017449999999999999,
|
|
"loss": 8.1418,
|
|
"mean_token_accuracy": 0.05746800377964974,
|
|
"num_tokens": 767236.0,
|
|
"step": 350
|
|
},
|
|
{
|
|
"entropy": 8.491841030120849,
|
|
"epoch": 0.3790710090763481,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.000177,
|
|
"loss": 8.1254,
|
|
"mean_token_accuracy": 0.06065751984715462,
|
|
"num_tokens": 778552.0,
|
|
"step": 355
|
|
},
|
|
{
|
|
"entropy": 8.428674793243408,
|
|
"epoch": 0.38441003737319807,
|
|
"grad_norm": 0.79296875,
|
|
"learning_rate": 0.0001795,
|
|
"loss": 8.1788,
|
|
"mean_token_accuracy": 0.05917464569211006,
|
|
"num_tokens": 789615.0,
|
|
"step": 360
|
|
},
|
|
{
|
|
"entropy": 8.48459882736206,
|
|
"epoch": 0.3897490656700481,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000182,
|
|
"loss": 8.1355,
|
|
"mean_token_accuracy": 0.06367055512964725,
|
|
"num_tokens": 799998.0,
|
|
"step": 365
|
|
},
|
|
{
|
|
"entropy": 8.322556686401366,
|
|
"epoch": 0.39508809396689804,
|
|
"grad_norm": 1.359375,
|
|
"learning_rate": 0.0001845,
|
|
"loss": 8.0679,
|
|
"mean_token_accuracy": 0.062298642471432686,
|
|
"num_tokens": 810042.0,
|
|
"step": 370
|
|
},
|
|
{
|
|
"entropy": 8.422898483276366,
|
|
"epoch": 0.400427122263748,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000187,
|
|
"loss": 8.0755,
|
|
"mean_token_accuracy": 0.05906784273684025,
|
|
"num_tokens": 821692.0,
|
|
"step": 375
|
|
},
|
|
{
|
|
"entropy": 8.36357364654541,
|
|
"epoch": 0.40576615056059795,
|
|
"grad_norm": 1.1484375,
|
|
"learning_rate": 0.0001895,
|
|
"loss": 8.1164,
|
|
"mean_token_accuracy": 0.06748287416994572,
|
|
"num_tokens": 832262.0,
|
|
"step": 380
|
|
},
|
|
{
|
|
"entropy": 8.333745193481445,
|
|
"epoch": 0.41110517885744796,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000192,
|
|
"loss": 8.04,
|
|
"mean_token_accuracy": 0.06151870898902416,
|
|
"num_tokens": 843006.0,
|
|
"step": 385
|
|
},
|
|
{
|
|
"entropy": 8.353145503997803,
|
|
"epoch": 0.4164442071542979,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0001945,
|
|
"loss": 8.0332,
|
|
"mean_token_accuracy": 0.062286530062556265,
|
|
"num_tokens": 853978.0,
|
|
"step": 390
|
|
},
|
|
{
|
|
"entropy": 8.357436656951904,
|
|
"epoch": 0.4217832354511479,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00019700000000000002,
|
|
"loss": 8.0743,
|
|
"mean_token_accuracy": 0.05977092795073986,
|
|
"num_tokens": 864634.0,
|
|
"step": 395
|
|
},
|
|
{
|
|
"entropy": 8.423241233825683,
|
|
"epoch": 0.4271222637479979,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.00019950000000000002,
|
|
"loss": 8.0568,
|
|
"mean_token_accuracy": 0.06662314645946026,
|
|
"num_tokens": 875379.0,
|
|
"step": 400
|
|
},
|
|
{
|
|
"entropy": 8.398690319061279,
|
|
"epoch": 0.43246129204484784,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.000202,
|
|
"loss": 8.0674,
|
|
"mean_token_accuracy": 0.06020735241472721,
|
|
"num_tokens": 886428.0,
|
|
"step": 405
|
|
},
|
|
{
|
|
"entropy": 8.425256729125977,
|
|
"epoch": 0.4378003203416978,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.00020449999999999998,
|
|
"loss": 8.0827,
|
|
"mean_token_accuracy": 0.0645828690379858,
|
|
"num_tokens": 896700.0,
|
|
"step": 410
|
|
},
|
|
{
|
|
"entropy": 8.369815444946289,
|
|
"epoch": 0.44313934863854776,
|
|
"grad_norm": 0.8125,
|
|
"learning_rate": 0.000207,
|
|
"loss": 8.0577,
|
|
"mean_token_accuracy": 0.06499057859182358,
|
|
"num_tokens": 907478.0,
|
|
"step": 415
|
|
},
|
|
{
|
|
"entropy": 8.39303207397461,
|
|
"epoch": 0.44847837693539777,
|
|
"grad_norm": 1.5234375,
|
|
"learning_rate": 0.0002095,
|
|
"loss": 7.9707,
|
|
"mean_token_accuracy": 0.07010989226400852,
|
|
"num_tokens": 918170.0,
|
|
"step": 420
|
|
},
|
|
{
|
|
"entropy": 8.28305778503418,
|
|
"epoch": 0.45381740523224773,
|
|
"grad_norm": 0.81640625,
|
|
"learning_rate": 0.000212,
|
|
"loss": 8.0345,
|
|
"mean_token_accuracy": 0.06685064136981964,
|
|
"num_tokens": 928866.0,
|
|
"step": 425
|
|
},
|
|
{
|
|
"entropy": 8.279892349243164,
|
|
"epoch": 0.4591564335290977,
|
|
"grad_norm": 0.80859375,
|
|
"learning_rate": 0.0002145,
|
|
"loss": 8.0205,
|
|
"mean_token_accuracy": 0.06419647410511971,
|
|
"num_tokens": 940395.0,
|
|
"step": 430
|
|
},
|
|
{
|
|
"entropy": 8.475063610076905,
|
|
"epoch": 0.4644954618259477,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.00021700000000000002,
|
|
"loss": 8.0365,
|
|
"mean_token_accuracy": 0.062472108751535416,
|
|
"num_tokens": 950814.0,
|
|
"step": 435
|
|
},
|
|
{
|
|
"entropy": 8.192297077178955,
|
|
"epoch": 0.46983449012279765,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0002195,
|
|
"loss": 8.0225,
|
|
"mean_token_accuracy": 0.0661535993218422,
|
|
"num_tokens": 961658.0,
|
|
"step": 440
|
|
},
|
|
{
|
|
"entropy": 8.43046703338623,
|
|
"epoch": 0.4751735184196476,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000222,
|
|
"loss": 8.0993,
|
|
"mean_token_accuracy": 0.06110986284911633,
|
|
"num_tokens": 974790.0,
|
|
"step": 445
|
|
},
|
|
{
|
|
"entropy": 8.331879615783691,
|
|
"epoch": 0.4805125467164976,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.0002245,
|
|
"loss": 7.9772,
|
|
"mean_token_accuracy": 0.06222324036061764,
|
|
"num_tokens": 986599.0,
|
|
"step": 450
|
|
},
|
|
{
|
|
"entropy": 8.243362045288086,
|
|
"epoch": 0.4858515750133476,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.00022700000000000002,
|
|
"loss": 7.9787,
|
|
"mean_token_accuracy": 0.06861450895667076,
|
|
"num_tokens": 997731.0,
|
|
"step": 455
|
|
},
|
|
{
|
|
"entropy": 8.290034008026122,
|
|
"epoch": 0.49119060331019754,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.00022950000000000002,
|
|
"loss": 8.0374,
|
|
"mean_token_accuracy": 0.06714494079351425,
|
|
"num_tokens": 1009343.0,
|
|
"step": 460
|
|
},
|
|
{
|
|
"entropy": 8.351651859283447,
|
|
"epoch": 0.4965296316070475,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.00023200000000000003,
|
|
"loss": 7.9541,
|
|
"mean_token_accuracy": 0.0648241501301527,
|
|
"num_tokens": 1019878.0,
|
|
"step": 465
|
|
},
|
|
{
|
|
"entropy": 8.248888206481933,
|
|
"epoch": 0.5018686599038975,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.00023449999999999998,
|
|
"loss": 7.9553,
|
|
"mean_token_accuracy": 0.07038948647677898,
|
|
"num_tokens": 1030551.0,
|
|
"step": 470
|
|
},
|
|
{
|
|
"entropy": 8.30935525894165,
|
|
"epoch": 0.5072076882007475,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.000237,
|
|
"loss": 7.9893,
|
|
"mean_token_accuracy": 0.06879530549049377,
|
|
"num_tokens": 1041432.0,
|
|
"step": 475
|
|
},
|
|
{
|
|
"entropy": 8.283312702178955,
|
|
"epoch": 0.5125467164975974,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0002395,
|
|
"loss": 7.9704,
|
|
"mean_token_accuracy": 0.06568947955965995,
|
|
"num_tokens": 1051769.0,
|
|
"step": 480
|
|
},
|
|
{
|
|
"entropy": 8.256877326965332,
|
|
"epoch": 0.5178857447944474,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000242,
|
|
"loss": 7.9282,
|
|
"mean_token_accuracy": 0.06400737687945365,
|
|
"num_tokens": 1062334.0,
|
|
"step": 485
|
|
},
|
|
{
|
|
"entropy": 8.27625093460083,
|
|
"epoch": 0.5232247730912973,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0002445,
|
|
"loss": 7.921,
|
|
"mean_token_accuracy": 0.06769514791667461,
|
|
"num_tokens": 1072792.0,
|
|
"step": 490
|
|
},
|
|
{
|
|
"entropy": 8.173498821258544,
|
|
"epoch": 0.5285638013881474,
|
|
"grad_norm": 1.203125,
|
|
"learning_rate": 0.000247,
|
|
"loss": 7.8148,
|
|
"mean_token_accuracy": 0.07047795243561268,
|
|
"num_tokens": 1083642.0,
|
|
"step": 495
|
|
},
|
|
{
|
|
"entropy": 8.192062282562256,
|
|
"epoch": 0.5339028296849974,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0002495,
|
|
"loss": 7.9111,
|
|
"mean_token_accuracy": 0.07375151067972183,
|
|
"num_tokens": 1094454.0,
|
|
"step": 500
|
|
},
|
|
{
|
|
"epoch": 0.5339028296849974,
|
|
"eval_entropy": 8.031556909734553,
|
|
"eval_loss": 8.007415771484375,
|
|
"eval_mean_token_accuracy": 0.06165650523715205,
|
|
"eval_num_tokens": 1094454.0,
|
|
"eval_runtime": 1.8278,
|
|
"eval_samples_per_second": 1685.073,
|
|
"eval_steps_per_second": 210.634,
|
|
"step": 500
|
|
},
|
|
{
|
|
"entropy": 8.316486740112305,
|
|
"epoch": 0.5392418579818473,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000252,
|
|
"loss": 7.9525,
|
|
"mean_token_accuracy": 0.06484691724181176,
|
|
"num_tokens": 1105068.0,
|
|
"step": 505
|
|
},
|
|
{
|
|
"entropy": 8.111512756347656,
|
|
"epoch": 0.5445808862786973,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0002545,
|
|
"loss": 7.9097,
|
|
"mean_token_accuracy": 0.07020260021090508,
|
|
"num_tokens": 1115668.0,
|
|
"step": 510
|
|
},
|
|
{
|
|
"entropy": 8.238842296600343,
|
|
"epoch": 0.5499199145755472,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000257,
|
|
"loss": 7.8867,
|
|
"mean_token_accuracy": 0.07027992233633995,
|
|
"num_tokens": 1126742.0,
|
|
"step": 515
|
|
},
|
|
{
|
|
"entropy": 8.179580593109131,
|
|
"epoch": 0.5552589428723972,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0002595,
|
|
"loss": 7.9668,
|
|
"mean_token_accuracy": 0.0645244125276804,
|
|
"num_tokens": 1137351.0,
|
|
"step": 520
|
|
},
|
|
{
|
|
"entropy": 8.22506046295166,
|
|
"epoch": 0.5605979711692471,
|
|
"grad_norm": 0.7421875,
|
|
"learning_rate": 0.000262,
|
|
"loss": 7.8967,
|
|
"mean_token_accuracy": 0.0711884018033743,
|
|
"num_tokens": 1148389.0,
|
|
"step": 525
|
|
},
|
|
{
|
|
"entropy": 8.167451763153077,
|
|
"epoch": 0.5659369994660972,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00026450000000000003,
|
|
"loss": 7.8994,
|
|
"mean_token_accuracy": 0.0703209575265646,
|
|
"num_tokens": 1158798.0,
|
|
"step": 530
|
|
},
|
|
{
|
|
"entropy": 8.239407634735107,
|
|
"epoch": 0.5712760277629472,
|
|
"grad_norm": 0.8203125,
|
|
"learning_rate": 0.00026700000000000004,
|
|
"loss": 7.948,
|
|
"mean_token_accuracy": 0.06852544359862804,
|
|
"num_tokens": 1169005.0,
|
|
"step": 535
|
|
},
|
|
{
|
|
"entropy": 8.212612915039063,
|
|
"epoch": 0.5766150560597971,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.00026950000000000005,
|
|
"loss": 7.8255,
|
|
"mean_token_accuracy": 0.06702273786067962,
|
|
"num_tokens": 1180560.0,
|
|
"step": 540
|
|
},
|
|
{
|
|
"entropy": 8.105984878540038,
|
|
"epoch": 0.5819540843566471,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.00027200000000000005,
|
|
"loss": 7.8035,
|
|
"mean_token_accuracy": 0.07094659134745598,
|
|
"num_tokens": 1191162.0,
|
|
"step": 545
|
|
},
|
|
{
|
|
"entropy": 8.162831687927246,
|
|
"epoch": 0.587293112653497,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0002745,
|
|
"loss": 7.8252,
|
|
"mean_token_accuracy": 0.06873422190546989,
|
|
"num_tokens": 1202017.0,
|
|
"step": 550
|
|
},
|
|
{
|
|
"entropy": 8.132958889007568,
|
|
"epoch": 0.592632140950347,
|
|
"grad_norm": 1.21875,
|
|
"learning_rate": 0.000277,
|
|
"loss": 7.8232,
|
|
"mean_token_accuracy": 0.07796628661453724,
|
|
"num_tokens": 1213356.0,
|
|
"step": 555
|
|
},
|
|
{
|
|
"entropy": 8.137973403930664,
|
|
"epoch": 0.5979711692471971,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0002795,
|
|
"loss": 7.8335,
|
|
"mean_token_accuracy": 0.07087706178426742,
|
|
"num_tokens": 1224164.0,
|
|
"step": 560
|
|
},
|
|
{
|
|
"entropy": 8.08298101425171,
|
|
"epoch": 0.603310197544047,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.00028199999999999997,
|
|
"loss": 7.8463,
|
|
"mean_token_accuracy": 0.0711651612073183,
|
|
"num_tokens": 1235483.0,
|
|
"step": 565
|
|
},
|
|
{
|
|
"entropy": 8.127214622497558,
|
|
"epoch": 0.608649225840897,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0002845,
|
|
"loss": 7.8595,
|
|
"mean_token_accuracy": 0.07112185694277287,
|
|
"num_tokens": 1248013.0,
|
|
"step": 570
|
|
},
|
|
{
|
|
"entropy": 8.093348598480224,
|
|
"epoch": 0.6139882541377469,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000287,
|
|
"loss": 7.7999,
|
|
"mean_token_accuracy": 0.07371774725615979,
|
|
"num_tokens": 1258512.0,
|
|
"step": 575
|
|
},
|
|
{
|
|
"entropy": 8.157558727264405,
|
|
"epoch": 0.6193272824345969,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0002895,
|
|
"loss": 7.8256,
|
|
"mean_token_accuracy": 0.06662675254046917,
|
|
"num_tokens": 1269425.0,
|
|
"step": 580
|
|
},
|
|
{
|
|
"entropy": 7.97422046661377,
|
|
"epoch": 0.6246663107314468,
|
|
"grad_norm": 0.7734375,
|
|
"learning_rate": 0.000292,
|
|
"loss": 7.7958,
|
|
"mean_token_accuracy": 0.07273220643401146,
|
|
"num_tokens": 1280595.0,
|
|
"step": 585
|
|
},
|
|
{
|
|
"entropy": 8.247371101379395,
|
|
"epoch": 0.6300053390282968,
|
|
"grad_norm": 1.1171875,
|
|
"learning_rate": 0.0002945,
|
|
"loss": 7.8485,
|
|
"mean_token_accuracy": 0.06550373584032058,
|
|
"num_tokens": 1290873.0,
|
|
"step": 590
|
|
},
|
|
{
|
|
"entropy": 8.015791416168213,
|
|
"epoch": 0.6353443673251469,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000297,
|
|
"loss": 7.852,
|
|
"mean_token_accuracy": 0.066574127972126,
|
|
"num_tokens": 1303150.0,
|
|
"step": 595
|
|
},
|
|
{
|
|
"entropy": 8.053708696365357,
|
|
"epoch": 0.6406833956219968,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.0002995,
|
|
"loss": 7.768,
|
|
"mean_token_accuracy": 0.07012191005051135,
|
|
"num_tokens": 1314862.0,
|
|
"step": 600
|
|
},
|
|
{
|
|
"entropy": 8.02850046157837,
|
|
"epoch": 0.6460224239188468,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.000302,
|
|
"loss": 7.8029,
|
|
"mean_token_accuracy": 0.07391900308430195,
|
|
"num_tokens": 1326052.0,
|
|
"step": 605
|
|
},
|
|
{
|
|
"entropy": 8.128786611557008,
|
|
"epoch": 0.6513614522156967,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0003045,
|
|
"loss": 7.7667,
|
|
"mean_token_accuracy": 0.0753765020519495,
|
|
"num_tokens": 1336847.0,
|
|
"step": 610
|
|
},
|
|
{
|
|
"entropy": 8.02261905670166,
|
|
"epoch": 0.6567004805125467,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.000307,
|
|
"loss": 7.7701,
|
|
"mean_token_accuracy": 0.06919390633702278,
|
|
"num_tokens": 1347611.0,
|
|
"step": 615
|
|
},
|
|
{
|
|
"entropy": 8.006579780578614,
|
|
"epoch": 0.6620395088093967,
|
|
"grad_norm": 1.3046875,
|
|
"learning_rate": 0.0003095,
|
|
"loss": 7.7859,
|
|
"mean_token_accuracy": 0.07407167293131352,
|
|
"num_tokens": 1358679.0,
|
|
"step": 620
|
|
},
|
|
{
|
|
"entropy": 8.048807954788208,
|
|
"epoch": 0.6673785371062466,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000312,
|
|
"loss": 7.7849,
|
|
"mean_token_accuracy": 0.07473289892077446,
|
|
"num_tokens": 1369409.0,
|
|
"step": 625
|
|
},
|
|
{
|
|
"entropy": 8.044457149505615,
|
|
"epoch": 0.6727175654030967,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0003145,
|
|
"loss": 7.7618,
|
|
"mean_token_accuracy": 0.06960076540708542,
|
|
"num_tokens": 1380380.0,
|
|
"step": 630
|
|
},
|
|
{
|
|
"entropy": 8.039245176315308,
|
|
"epoch": 0.6780565936999466,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000317,
|
|
"loss": 7.7081,
|
|
"mean_token_accuracy": 0.07184483371675014,
|
|
"num_tokens": 1390885.0,
|
|
"step": 635
|
|
},
|
|
{
|
|
"entropy": 7.9913702487945555,
|
|
"epoch": 0.6833956219967966,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0003195,
|
|
"loss": 7.7418,
|
|
"mean_token_accuracy": 0.07358726263046264,
|
|
"num_tokens": 1400960.0,
|
|
"step": 640
|
|
},
|
|
{
|
|
"entropy": 7.986459016799927,
|
|
"epoch": 0.6887346502936466,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000322,
|
|
"loss": 7.6917,
|
|
"mean_token_accuracy": 0.07739836499094963,
|
|
"num_tokens": 1411484.0,
|
|
"step": 645
|
|
},
|
|
{
|
|
"entropy": 8.129097318649292,
|
|
"epoch": 0.6940736785904965,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.00032450000000000003,
|
|
"loss": 7.7908,
|
|
"mean_token_accuracy": 0.07160034552216529,
|
|
"num_tokens": 1422931.0,
|
|
"step": 650
|
|
},
|
|
{
|
|
"entropy": 7.91273045539856,
|
|
"epoch": 0.6994127068873465,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.00032700000000000003,
|
|
"loss": 7.7237,
|
|
"mean_token_accuracy": 0.07407537326216698,
|
|
"num_tokens": 1434347.0,
|
|
"step": 655
|
|
},
|
|
{
|
|
"entropy": 8.050477170944214,
|
|
"epoch": 0.7047517351841964,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00032950000000000004,
|
|
"loss": 7.8135,
|
|
"mean_token_accuracy": 0.07437525913119317,
|
|
"num_tokens": 1445687.0,
|
|
"step": 660
|
|
},
|
|
{
|
|
"entropy": 8.082316970825195,
|
|
"epoch": 0.7100907634810465,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.00033200000000000005,
|
|
"loss": 7.6471,
|
|
"mean_token_accuracy": 0.08003439009189606,
|
|
"num_tokens": 1456335.0,
|
|
"step": 665
|
|
},
|
|
{
|
|
"entropy": 7.9566230297088625,
|
|
"epoch": 0.7154297917778965,
|
|
"grad_norm": 1.1953125,
|
|
"learning_rate": 0.00033450000000000005,
|
|
"loss": 7.7071,
|
|
"mean_token_accuracy": 0.07267946116626263,
|
|
"num_tokens": 1467369.0,
|
|
"step": 670
|
|
},
|
|
{
|
|
"entropy": 7.986024713516235,
|
|
"epoch": 0.7207688200747464,
|
|
"grad_norm": 1.171875,
|
|
"learning_rate": 0.000337,
|
|
"loss": 7.7612,
|
|
"mean_token_accuracy": 0.06941658817231655,
|
|
"num_tokens": 1478734.0,
|
|
"step": 675
|
|
},
|
|
{
|
|
"entropy": 7.984552240371704,
|
|
"epoch": 0.7261078483715964,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0003395,
|
|
"loss": 7.7359,
|
|
"mean_token_accuracy": 0.07321545220911503,
|
|
"num_tokens": 1489807.0,
|
|
"step": 680
|
|
},
|
|
{
|
|
"entropy": 7.967883205413818,
|
|
"epoch": 0.7314468766684463,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.000342,
|
|
"loss": 7.6794,
|
|
"mean_token_accuracy": 0.07043348215520381,
|
|
"num_tokens": 1501646.0,
|
|
"step": 685
|
|
},
|
|
{
|
|
"entropy": 7.948320388793945,
|
|
"epoch": 0.7367859049652963,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.00034449999999999997,
|
|
"loss": 7.6987,
|
|
"mean_token_accuracy": 0.07335374914109707,
|
|
"num_tokens": 1511849.0,
|
|
"step": 690
|
|
},
|
|
{
|
|
"entropy": 7.999287462234497,
|
|
"epoch": 0.7421249332621462,
|
|
"grad_norm": 1.3125,
|
|
"learning_rate": 0.000347,
|
|
"loss": 7.7004,
|
|
"mean_token_accuracy": 0.07678697407245635,
|
|
"num_tokens": 1522183.0,
|
|
"step": 695
|
|
},
|
|
{
|
|
"entropy": 7.895766305923462,
|
|
"epoch": 0.7474639615589963,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0003495,
|
|
"loss": 7.7063,
|
|
"mean_token_accuracy": 0.07557001523673534,
|
|
"num_tokens": 1532819.0,
|
|
"step": 700
|
|
},
|
|
{
|
|
"entropy": 8.02945156097412,
|
|
"epoch": 0.7528029898558463,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.000352,
|
|
"loss": 7.6718,
|
|
"mean_token_accuracy": 0.07557161971926689,
|
|
"num_tokens": 1543909.0,
|
|
"step": 705
|
|
},
|
|
{
|
|
"entropy": 7.9088582515716555,
|
|
"epoch": 0.7581420181526962,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0003545,
|
|
"loss": 7.7087,
|
|
"mean_token_accuracy": 0.07322404161095619,
|
|
"num_tokens": 1554147.0,
|
|
"step": 710
|
|
},
|
|
{
|
|
"entropy": 7.878606176376342,
|
|
"epoch": 0.7634810464495462,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000357,
|
|
"loss": 7.6136,
|
|
"mean_token_accuracy": 0.07956673577427864,
|
|
"num_tokens": 1565466.0,
|
|
"step": 715
|
|
},
|
|
{
|
|
"entropy": 7.8776044845581055,
|
|
"epoch": 0.7688200747463961,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0003595,
|
|
"loss": 7.5815,
|
|
"mean_token_accuracy": 0.07377611920237541,
|
|
"num_tokens": 1577264.0,
|
|
"step": 720
|
|
},
|
|
{
|
|
"entropy": 7.885418844223023,
|
|
"epoch": 0.7741591030432461,
|
|
"grad_norm": 1.1796875,
|
|
"learning_rate": 0.000362,
|
|
"loss": 7.686,
|
|
"mean_token_accuracy": 0.0757551722228527,
|
|
"num_tokens": 1588222.0,
|
|
"step": 725
|
|
},
|
|
{
|
|
"entropy": 7.901127099990845,
|
|
"epoch": 0.7794981313400962,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0003645,
|
|
"loss": 7.6224,
|
|
"mean_token_accuracy": 0.07221109680831432,
|
|
"num_tokens": 1600609.0,
|
|
"step": 730
|
|
},
|
|
{
|
|
"entropy": 7.929693412780762,
|
|
"epoch": 0.7848371596369461,
|
|
"grad_norm": 1.125,
|
|
"learning_rate": 0.000367,
|
|
"loss": 7.669,
|
|
"mean_token_accuracy": 0.07948594614863395,
|
|
"num_tokens": 1612032.0,
|
|
"step": 735
|
|
},
|
|
{
|
|
"entropy": 7.876828384399414,
|
|
"epoch": 0.7901761879337961,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0003695,
|
|
"loss": 7.6973,
|
|
"mean_token_accuracy": 0.0742558129131794,
|
|
"num_tokens": 1623228.0,
|
|
"step": 740
|
|
},
|
|
{
|
|
"entropy": 7.857063436508179,
|
|
"epoch": 0.795515216230646,
|
|
"grad_norm": 1.15625,
|
|
"learning_rate": 0.000372,
|
|
"loss": 7.6469,
|
|
"mean_token_accuracy": 0.07817403376102447,
|
|
"num_tokens": 1634293.0,
|
|
"step": 745
|
|
},
|
|
{
|
|
"entropy": 7.910915946960449,
|
|
"epoch": 0.800854244527496,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003745,
|
|
"loss": 7.6183,
|
|
"mean_token_accuracy": 0.08015101104974746,
|
|
"num_tokens": 1645782.0,
|
|
"step": 750
|
|
},
|
|
{
|
|
"entropy": 7.885407543182373,
|
|
"epoch": 0.8061932728243459,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000377,
|
|
"loss": 7.5871,
|
|
"mean_token_accuracy": 0.0775064155459404,
|
|
"num_tokens": 1657060.0,
|
|
"step": 755
|
|
},
|
|
{
|
|
"entropy": 7.78002610206604,
|
|
"epoch": 0.8115323011211959,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0003795,
|
|
"loss": 7.568,
|
|
"mean_token_accuracy": 0.08284709006547927,
|
|
"num_tokens": 1668017.0,
|
|
"step": 760
|
|
},
|
|
{
|
|
"entropy": 7.908759117126465,
|
|
"epoch": 0.816871329418046,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000382,
|
|
"loss": 7.6398,
|
|
"mean_token_accuracy": 0.07324047535657882,
|
|
"num_tokens": 1678331.0,
|
|
"step": 765
|
|
},
|
|
{
|
|
"entropy": 7.884358358383179,
|
|
"epoch": 0.8222103577148959,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.0003845,
|
|
"loss": 7.5591,
|
|
"mean_token_accuracy": 0.0813971921801567,
|
|
"num_tokens": 1688719.0,
|
|
"step": 770
|
|
},
|
|
{
|
|
"entropy": 7.755726432800293,
|
|
"epoch": 0.8275493860117459,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00038700000000000003,
|
|
"loss": 7.5821,
|
|
"mean_token_accuracy": 0.07562258839607239,
|
|
"num_tokens": 1699028.0,
|
|
"step": 775
|
|
},
|
|
{
|
|
"entropy": 7.950335073471069,
|
|
"epoch": 0.8328884143085958,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00038950000000000003,
|
|
"loss": 7.5798,
|
|
"mean_token_accuracy": 0.07690282315015792,
|
|
"num_tokens": 1711122.0,
|
|
"step": 780
|
|
},
|
|
{
|
|
"entropy": 7.790361976623535,
|
|
"epoch": 0.8382274426054458,
|
|
"grad_norm": 1.109375,
|
|
"learning_rate": 0.00039200000000000004,
|
|
"loss": 7.5939,
|
|
"mean_token_accuracy": 0.07726500257849693,
|
|
"num_tokens": 1721694.0,
|
|
"step": 785
|
|
},
|
|
{
|
|
"entropy": 7.733095169067383,
|
|
"epoch": 0.8435664709022958,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.00039450000000000005,
|
|
"loss": 7.5939,
|
|
"mean_token_accuracy": 0.0780683733522892,
|
|
"num_tokens": 1732530.0,
|
|
"step": 790
|
|
},
|
|
{
|
|
"entropy": 7.877895975112915,
|
|
"epoch": 0.8489054991991457,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00039700000000000005,
|
|
"loss": 7.5577,
|
|
"mean_token_accuracy": 0.08093490153551101,
|
|
"num_tokens": 1742860.0,
|
|
"step": 795
|
|
},
|
|
{
|
|
"entropy": 7.750740194320679,
|
|
"epoch": 0.8542445274959958,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0003995,
|
|
"loss": 7.5315,
|
|
"mean_token_accuracy": 0.0812942422926426,
|
|
"num_tokens": 1752900.0,
|
|
"step": 800
|
|
},
|
|
{
|
|
"entropy": 7.801487588882447,
|
|
"epoch": 0.8595835557928457,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000402,
|
|
"loss": 7.5465,
|
|
"mean_token_accuracy": 0.07882912531495094,
|
|
"num_tokens": 1763293.0,
|
|
"step": 805
|
|
},
|
|
{
|
|
"entropy": 7.752074146270752,
|
|
"epoch": 0.8649225840896957,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004045,
|
|
"loss": 7.5696,
|
|
"mean_token_accuracy": 0.0819572426378727,
|
|
"num_tokens": 1773893.0,
|
|
"step": 810
|
|
},
|
|
{
|
|
"entropy": 7.815144634246826,
|
|
"epoch": 0.8702616123865456,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.00040699999999999997,
|
|
"loss": 7.6358,
|
|
"mean_token_accuracy": 0.07636146545410157,
|
|
"num_tokens": 1785568.0,
|
|
"step": 815
|
|
},
|
|
{
|
|
"entropy": 7.8184651851654055,
|
|
"epoch": 0.8756006406833956,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004095,
|
|
"loss": 7.5594,
|
|
"mean_token_accuracy": 0.08072886392474174,
|
|
"num_tokens": 1797387.0,
|
|
"step": 820
|
|
},
|
|
{
|
|
"entropy": 7.813911390304566,
|
|
"epoch": 0.8809396689802456,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.000412,
|
|
"loss": 7.5902,
|
|
"mean_token_accuracy": 0.07879998087882996,
|
|
"num_tokens": 1808382.0,
|
|
"step": 825
|
|
},
|
|
{
|
|
"entropy": 7.829354333877563,
|
|
"epoch": 0.8862786972770955,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004145,
|
|
"loss": 7.5745,
|
|
"mean_token_accuracy": 0.0824002705514431,
|
|
"num_tokens": 1818937.0,
|
|
"step": 830
|
|
},
|
|
{
|
|
"entropy": 7.799869441986084,
|
|
"epoch": 0.8916177255739456,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000417,
|
|
"loss": 7.5099,
|
|
"mean_token_accuracy": 0.07985455431044101,
|
|
"num_tokens": 1830547.0,
|
|
"step": 835
|
|
},
|
|
{
|
|
"entropy": 7.7153465270996096,
|
|
"epoch": 0.8969567538707955,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004195,
|
|
"loss": 7.5052,
|
|
"mean_token_accuracy": 0.08278301656246186,
|
|
"num_tokens": 1841569.0,
|
|
"step": 840
|
|
},
|
|
{
|
|
"entropy": 7.764185190200806,
|
|
"epoch": 0.9022957821676455,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000422,
|
|
"loss": 7.5154,
|
|
"mean_token_accuracy": 0.08583573549985886,
|
|
"num_tokens": 1852300.0,
|
|
"step": 845
|
|
},
|
|
{
|
|
"entropy": 7.718573999404907,
|
|
"epoch": 0.9076348104644955,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004245,
|
|
"loss": 7.5077,
|
|
"mean_token_accuracy": 0.08417970538139344,
|
|
"num_tokens": 1863664.0,
|
|
"step": 850
|
|
},
|
|
{
|
|
"entropy": 7.6863548278808596,
|
|
"epoch": 0.9129738387613454,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.000427,
|
|
"loss": 7.4426,
|
|
"mean_token_accuracy": 0.08229955136775971,
|
|
"num_tokens": 1873623.0,
|
|
"step": 855
|
|
},
|
|
{
|
|
"entropy": 7.687994432449341,
|
|
"epoch": 0.9183128670581954,
|
|
"grad_norm": 1.1328125,
|
|
"learning_rate": 0.0004295,
|
|
"loss": 7.5429,
|
|
"mean_token_accuracy": 0.0791929729282856,
|
|
"num_tokens": 1886092.0,
|
|
"step": 860
|
|
},
|
|
{
|
|
"entropy": 7.751056241989136,
|
|
"epoch": 0.9236518953550453,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000432,
|
|
"loss": 7.5525,
|
|
"mean_token_accuracy": 0.08293391354382038,
|
|
"num_tokens": 1897983.0,
|
|
"step": 865
|
|
},
|
|
{
|
|
"entropy": 7.769723224639892,
|
|
"epoch": 0.9289909236518954,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.0004345,
|
|
"loss": 7.5134,
|
|
"mean_token_accuracy": 0.08427821174263954,
|
|
"num_tokens": 1908193.0,
|
|
"step": 870
|
|
},
|
|
{
|
|
"entropy": 7.7698462963104244,
|
|
"epoch": 0.9343299519487454,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.000437,
|
|
"loss": 7.4921,
|
|
"mean_token_accuracy": 0.08160046115517616,
|
|
"num_tokens": 1918628.0,
|
|
"step": 875
|
|
},
|
|
{
|
|
"entropy": 7.668918085098267,
|
|
"epoch": 0.9396689802455953,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.0004395,
|
|
"loss": 7.5372,
|
|
"mean_token_accuracy": 0.08014441058039665,
|
|
"num_tokens": 1929071.0,
|
|
"step": 880
|
|
},
|
|
{
|
|
"entropy": 7.654262733459473,
|
|
"epoch": 0.9450080085424453,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.000442,
|
|
"loss": 7.5003,
|
|
"mean_token_accuracy": 0.08210862651467324,
|
|
"num_tokens": 1939567.0,
|
|
"step": 885
|
|
},
|
|
{
|
|
"entropy": 7.773276090621948,
|
|
"epoch": 0.9503470368392952,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004445,
|
|
"loss": 7.5402,
|
|
"mean_token_accuracy": 0.08018058091402054,
|
|
"num_tokens": 1951165.0,
|
|
"step": 890
|
|
},
|
|
{
|
|
"entropy": 7.798144149780273,
|
|
"epoch": 0.9556860651361452,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.000447,
|
|
"loss": 7.4594,
|
|
"mean_token_accuracy": 0.08885749988257885,
|
|
"num_tokens": 1960855.0,
|
|
"step": 895
|
|
},
|
|
{
|
|
"entropy": 7.707589864730835,
|
|
"epoch": 0.9610250934329952,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.00044950000000000003,
|
|
"loss": 7.4564,
|
|
"mean_token_accuracy": 0.08240535110235214,
|
|
"num_tokens": 1971873.0,
|
|
"step": 900
|
|
},
|
|
{
|
|
"entropy": 7.671180057525635,
|
|
"epoch": 0.9663641217298452,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.00045200000000000004,
|
|
"loss": 7.4848,
|
|
"mean_token_accuracy": 0.08260536044836045,
|
|
"num_tokens": 1983561.0,
|
|
"step": 905
|
|
},
|
|
{
|
|
"entropy": 7.731744003295899,
|
|
"epoch": 0.9717031500266952,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.00045450000000000004,
|
|
"loss": 7.4828,
|
|
"mean_token_accuracy": 0.08393343463540077,
|
|
"num_tokens": 1994534.0,
|
|
"step": 910
|
|
},
|
|
{
|
|
"entropy": 7.590392780303955,
|
|
"epoch": 0.9770421783235451,
|
|
"grad_norm": 1.09375,
|
|
"learning_rate": 0.00045700000000000005,
|
|
"loss": 7.432,
|
|
"mean_token_accuracy": 0.08598880358040333,
|
|
"num_tokens": 2005121.0,
|
|
"step": 915
|
|
},
|
|
{
|
|
"entropy": 7.696979951858521,
|
|
"epoch": 0.9823812066203951,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.00045950000000000006,
|
|
"loss": 7.4211,
|
|
"mean_token_accuracy": 0.08999745771288872,
|
|
"num_tokens": 2016260.0,
|
|
"step": 920
|
|
},
|
|
{
|
|
"entropy": 7.615106248855591,
|
|
"epoch": 0.987720234917245,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.000462,
|
|
"loss": 7.4074,
|
|
"mean_token_accuracy": 0.08892037123441696,
|
|
"num_tokens": 2025705.0,
|
|
"step": 925
|
|
},
|
|
{
|
|
"entropy": 7.7097069263458256,
|
|
"epoch": 0.993059263214095,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004645,
|
|
"loss": 7.5596,
|
|
"mean_token_accuracy": 0.07989178895950318,
|
|
"num_tokens": 2036675.0,
|
|
"step": 930
|
|
},
|
|
{
|
|
"entropy": 7.544618844985962,
|
|
"epoch": 0.9983982915109451,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000467,
|
|
"loss": 7.4114,
|
|
"mean_token_accuracy": 0.08617193698883056,
|
|
"num_tokens": 2047877.0,
|
|
"step": 935
|
|
},
|
|
{
|
|
"entropy": 7.8020609219868975,
|
|
"epoch": 1.0032034169781099,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004695,
|
|
"loss": 7.3332,
|
|
"mean_token_accuracy": 0.08596346527338028,
|
|
"num_tokens": 2057153.0,
|
|
"step": 940
|
|
},
|
|
{
|
|
"entropy": 7.623947048187256,
|
|
"epoch": 1.00854244527496,
|
|
"grad_norm": 1.265625,
|
|
"learning_rate": 0.000472,
|
|
"loss": 7.4568,
|
|
"mean_token_accuracy": 0.0771791972219944,
|
|
"num_tokens": 2068760.0,
|
|
"step": 945
|
|
},
|
|
{
|
|
"entropy": 7.589405059814453,
|
|
"epoch": 1.01388147357181,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004745,
|
|
"loss": 7.3104,
|
|
"mean_token_accuracy": 0.0916103646159172,
|
|
"num_tokens": 2078671.0,
|
|
"step": 950
|
|
},
|
|
{
|
|
"entropy": 7.588209199905395,
|
|
"epoch": 1.01922050186866,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.000477,
|
|
"loss": 7.3495,
|
|
"mean_token_accuracy": 0.08704460337758065,
|
|
"num_tokens": 2089581.0,
|
|
"step": 955
|
|
},
|
|
{
|
|
"entropy": 7.6823986053466795,
|
|
"epoch": 1.02455953016551,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004795,
|
|
"loss": 7.293,
|
|
"mean_token_accuracy": 0.09483219981193543,
|
|
"num_tokens": 2100026.0,
|
|
"step": 960
|
|
},
|
|
{
|
|
"entropy": 7.51586127281189,
|
|
"epoch": 1.02989855846236,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.000482,
|
|
"loss": 7.3062,
|
|
"mean_token_accuracy": 0.09631321728229522,
|
|
"num_tokens": 2111838.0,
|
|
"step": 965
|
|
},
|
|
{
|
|
"entropy": 7.603687572479248,
|
|
"epoch": 1.0352375867592098,
|
|
"grad_norm": 0.765625,
|
|
"learning_rate": 0.0004845,
|
|
"loss": 7.3908,
|
|
"mean_token_accuracy": 0.0855435535311699,
|
|
"num_tokens": 2123063.0,
|
|
"step": 970
|
|
},
|
|
{
|
|
"entropy": 7.658422660827637,
|
|
"epoch": 1.0405766150560598,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000487,
|
|
"loss": 7.279,
|
|
"mean_token_accuracy": 0.09217825904488564,
|
|
"num_tokens": 2134321.0,
|
|
"step": 975
|
|
},
|
|
{
|
|
"entropy": 7.596205329895019,
|
|
"epoch": 1.0459156433529098,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004895,
|
|
"loss": 7.3405,
|
|
"mean_token_accuracy": 0.09226287379860879,
|
|
"num_tokens": 2144216.0,
|
|
"step": 980
|
|
},
|
|
{
|
|
"entropy": 7.586561298370361,
|
|
"epoch": 1.0512546716497597,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.000492,
|
|
"loss": 7.3236,
|
|
"mean_token_accuracy": 0.08801514208316803,
|
|
"num_tokens": 2155757.0,
|
|
"step": 985
|
|
},
|
|
{
|
|
"entropy": 7.509102058410645,
|
|
"epoch": 1.0565936999466097,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004945,
|
|
"loss": 7.2762,
|
|
"mean_token_accuracy": 0.08867826834321021,
|
|
"num_tokens": 2166661.0,
|
|
"step": 990
|
|
},
|
|
{
|
|
"entropy": 7.554547595977783,
|
|
"epoch": 1.0619327282434596,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.000497,
|
|
"loss": 7.2522,
|
|
"mean_token_accuracy": 0.0905068688094616,
|
|
"num_tokens": 2176796.0,
|
|
"step": 995
|
|
},
|
|
{
|
|
"entropy": 7.565032482147217,
|
|
"epoch": 1.0672717565403096,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004995,
|
|
"loss": 7.3151,
|
|
"mean_token_accuracy": 0.09214983209967613,
|
|
"num_tokens": 2186821.0,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"epoch": 1.0672717565403096,
|
|
"eval_entropy": 7.661522679514699,
|
|
"eval_loss": 7.484097957611084,
|
|
"eval_mean_token_accuracy": 0.08444042449163926,
|
|
"eval_num_tokens": 2186821.0,
|
|
"eval_runtime": 1.8185,
|
|
"eval_samples_per_second": 1693.67,
|
|
"eval_steps_per_second": 211.709,
|
|
"step": 1000
|
|
},
|
|
{
|
|
"entropy": 7.488584995269775,
|
|
"epoch": 1.0726107848371595,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004999997458093253,
|
|
"loss": 7.2872,
|
|
"mean_token_accuracy": 0.08735908269882202,
|
|
"num_tokens": 2197953.0,
|
|
"step": 1005
|
|
},
|
|
{
|
|
"entropy": 7.587278175354004,
|
|
"epoch": 1.0779498131340097,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004999987131606938,
|
|
"loss": 7.2946,
|
|
"mean_token_accuracy": 0.09318367019295692,
|
|
"num_tokens": 2208033.0,
|
|
"step": 1010
|
|
},
|
|
{
|
|
"entropy": 7.508203601837158,
|
|
"epoch": 1.0832888414308597,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004999968861708311,
|
|
"loss": 7.306,
|
|
"mean_token_accuracy": 0.0833747498691082,
|
|
"num_tokens": 2218010.0,
|
|
"step": 1015
|
|
},
|
|
{
|
|
"entropy": 7.607468748092652,
|
|
"epoch": 1.0886278697277096,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004999942648461873,
|
|
"loss": 7.3213,
|
|
"mean_token_accuracy": 0.08709867373108864,
|
|
"num_tokens": 2229079.0,
|
|
"step": 1020
|
|
},
|
|
{
|
|
"entropy": 7.554487895965576,
|
|
"epoch": 1.0939668980245596,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004999908491960169,
|
|
"loss": 7.3188,
|
|
"mean_token_accuracy": 0.08554268255829811,
|
|
"num_tokens": 2240333.0,
|
|
"step": 1025
|
|
},
|
|
{
|
|
"entropy": 7.41667366027832,
|
|
"epoch": 1.0993059263214096,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004999866392323784,
|
|
"loss": 7.2619,
|
|
"mean_token_accuracy": 0.09279655292630196,
|
|
"num_tokens": 2251385.0,
|
|
"step": 1030
|
|
},
|
|
{
|
|
"entropy": 7.586219072341919,
|
|
"epoch": 1.1046449546182595,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004999816349701349,
|
|
"loss": 7.2562,
|
|
"mean_token_accuracy": 0.09476209431886673,
|
|
"num_tokens": 2261890.0,
|
|
"step": 1035
|
|
},
|
|
{
|
|
"entropy": 7.429596185684204,
|
|
"epoch": 1.1099839829151095,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004999758364269535,
|
|
"loss": 7.287,
|
|
"mean_token_accuracy": 0.0902016930282116,
|
|
"num_tokens": 2272510.0,
|
|
"step": 1040
|
|
},
|
|
{
|
|
"entropy": 7.436788702011109,
|
|
"epoch": 1.1153230112119594,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004999692436233055,
|
|
"loss": 7.2791,
|
|
"mean_token_accuracy": 0.08760200515389442,
|
|
"num_tokens": 2283392.0,
|
|
"step": 1045
|
|
},
|
|
{
|
|
"entropy": 7.570240783691406,
|
|
"epoch": 1.1206620395088094,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004999618565824665,
|
|
"loss": 7.2279,
|
|
"mean_token_accuracy": 0.09378096982836723,
|
|
"num_tokens": 2295155.0,
|
|
"step": 1050
|
|
},
|
|
{
|
|
"entropy": 7.490195608139038,
|
|
"epoch": 1.1260010678056593,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999536753305157,
|
|
"loss": 7.3268,
|
|
"mean_token_accuracy": 0.08983222171664237,
|
|
"num_tokens": 2307208.0,
|
|
"step": 1055
|
|
},
|
|
{
|
|
"entropy": 7.629471921920777,
|
|
"epoch": 1.1313400961025093,
|
|
"grad_norm": 1.046875,
|
|
"learning_rate": 0.0004999446998963364,
|
|
"loss": 7.2481,
|
|
"mean_token_accuracy": 0.09013928584754467,
|
|
"num_tokens": 2317480.0,
|
|
"step": 1060
|
|
},
|
|
{
|
|
"entropy": 7.449492406845093,
|
|
"epoch": 1.1366791243993593,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004999349303116159,
|
|
"loss": 7.3373,
|
|
"mean_token_accuracy": 0.08976641297340393,
|
|
"num_tokens": 2328840.0,
|
|
"step": 1065
|
|
},
|
|
{
|
|
"entropy": 7.580901098251343,
|
|
"epoch": 1.1420181526962092,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004999243666108448,
|
|
"loss": 7.285,
|
|
"mean_token_accuracy": 0.09045098423957824,
|
|
"num_tokens": 2340450.0,
|
|
"step": 1070
|
|
},
|
|
{
|
|
"entropy": 7.4406897068023685,
|
|
"epoch": 1.1473571809930592,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004999130088313176,
|
|
"loss": 7.3033,
|
|
"mean_token_accuracy": 0.09590284526348114,
|
|
"num_tokens": 2351415.0,
|
|
"step": 1075
|
|
},
|
|
{
|
|
"entropy": 7.485285329818725,
|
|
"epoch": 1.1526962092899091,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004999008570131319,
|
|
"loss": 7.2965,
|
|
"mean_token_accuracy": 0.09630362093448638,
|
|
"num_tokens": 2362153.0,
|
|
"step": 1080
|
|
},
|
|
{
|
|
"entropy": 7.625531721115112,
|
|
"epoch": 1.1580352375867593,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004998879111991889,
|
|
"loss": 7.2759,
|
|
"mean_token_accuracy": 0.089786946028471,
|
|
"num_tokens": 2372352.0,
|
|
"step": 1085
|
|
},
|
|
{
|
|
"entropy": 7.374940299987793,
|
|
"epoch": 1.1633742658836093,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000499874171435193,
|
|
"loss": 7.2659,
|
|
"mean_token_accuracy": 0.09284015148878097,
|
|
"num_tokens": 2382294.0,
|
|
"step": 1090
|
|
},
|
|
{
|
|
"entropy": 7.572332715988159,
|
|
"epoch": 1.1687132941804592,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004998596377696511,
|
|
"loss": 7.2382,
|
|
"mean_token_accuracy": 0.09612747579813004,
|
|
"num_tokens": 2394054.0,
|
|
"step": 1095
|
|
},
|
|
{
|
|
"entropy": 7.367642450332641,
|
|
"epoch": 1.1740523224773092,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004998443102538734,
|
|
"loss": 7.2585,
|
|
"mean_token_accuracy": 0.09028631299734116,
|
|
"num_tokens": 2405408.0,
|
|
"step": 1100
|
|
},
|
|
{
|
|
"entropy": 7.572077369689941,
|
|
"epoch": 1.1793913507741591,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004998281889419725,
|
|
"loss": 7.2905,
|
|
"mean_token_accuracy": 0.08611830845475196,
|
|
"num_tokens": 2416378.0,
|
|
"step": 1105
|
|
},
|
|
{
|
|
"entropy": 7.381543397903442,
|
|
"epoch": 1.184730379071009,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004998112738908636,
|
|
"loss": 7.183,
|
|
"mean_token_accuracy": 0.09558979794383049,
|
|
"num_tokens": 2427016.0,
|
|
"step": 1110
|
|
},
|
|
{
|
|
"entropy": 7.501509714126587,
|
|
"epoch": 1.190069407367859,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004997935651602641,
|
|
"loss": 7.1302,
|
|
"mean_token_accuracy": 0.09728729203343392,
|
|
"num_tokens": 2437824.0,
|
|
"step": 1115
|
|
},
|
|
{
|
|
"entropy": 7.4092631340026855,
|
|
"epoch": 1.195408435664709,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004997750628126931,
|
|
"loss": 7.2588,
|
|
"mean_token_accuracy": 0.08902693465352059,
|
|
"num_tokens": 2448829.0,
|
|
"step": 1120
|
|
},
|
|
{
|
|
"entropy": 7.444784927368164,
|
|
"epoch": 1.200747463961559,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004997557669134718,
|
|
"loss": 7.2186,
|
|
"mean_token_accuracy": 0.09314063340425491,
|
|
"num_tokens": 2459629.0,
|
|
"step": 1125
|
|
},
|
|
{
|
|
"entropy": 7.362234878540039,
|
|
"epoch": 1.206086492258409,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004997356775307231,
|
|
"loss": 7.2803,
|
|
"mean_token_accuracy": 0.09419681504368782,
|
|
"num_tokens": 2471478.0,
|
|
"step": 1130
|
|
},
|
|
{
|
|
"entropy": 7.532480144500733,
|
|
"epoch": 1.2114255205552589,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.000499714794735371,
|
|
"loss": 7.1266,
|
|
"mean_token_accuracy": 0.10543641149997711,
|
|
"num_tokens": 2481573.0,
|
|
"step": 1135
|
|
},
|
|
{
|
|
"entropy": 7.350391530990601,
|
|
"epoch": 1.2167645488521088,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004996931186011406,
|
|
"loss": 7.256,
|
|
"mean_token_accuracy": 0.09152917265892029,
|
|
"num_tokens": 2493429.0,
|
|
"step": 1140
|
|
},
|
|
{
|
|
"entropy": 7.421112728118897,
|
|
"epoch": 1.222103577148959,
|
|
"grad_norm": 1.0546875,
|
|
"learning_rate": 0.000499670649204558,
|
|
"loss": 7.218,
|
|
"mean_token_accuracy": 0.09748484790325165,
|
|
"num_tokens": 2503882.0,
|
|
"step": 1145
|
|
},
|
|
{
|
|
"entropy": 7.419456720352173,
|
|
"epoch": 1.227442605445809,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004996473866249496,
|
|
"loss": 7.1971,
|
|
"mean_token_accuracy": 0.09453025832772255,
|
|
"num_tokens": 2513978.0,
|
|
"step": 1150
|
|
},
|
|
{
|
|
"entropy": 7.512584495544433,
|
|
"epoch": 1.232781633742659,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004996233309444424,
|
|
"loss": 7.2582,
|
|
"mean_token_accuracy": 0.09497607797384262,
|
|
"num_tokens": 2523806.0,
|
|
"step": 1155
|
|
},
|
|
{
|
|
"entropy": 7.397188282012939,
|
|
"epoch": 1.2381206620395089,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.000499598482247963,
|
|
"loss": 7.2416,
|
|
"mean_token_accuracy": 0.09581049084663391,
|
|
"num_tokens": 2533695.0,
|
|
"step": 1160
|
|
},
|
|
{
|
|
"entropy": 7.449572372436523,
|
|
"epoch": 1.2434596903363588,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004995728406232378,
|
|
"loss": 7.1616,
|
|
"mean_token_accuracy": 0.10089872553944587,
|
|
"num_tokens": 2544315.0,
|
|
"step": 1165
|
|
},
|
|
{
|
|
"entropy": 7.433754110336304,
|
|
"epoch": 1.2487987186332088,
|
|
"grad_norm": 1.2421875,
|
|
"learning_rate": 0.000499546406160793,
|
|
"loss": 7.1648,
|
|
"mean_token_accuracy": 0.09890817850828171,
|
|
"num_tokens": 2555985.0,
|
|
"step": 1170
|
|
},
|
|
{
|
|
"entropy": 7.336302137374878,
|
|
"epoch": 1.2541377469300587,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004995191789539531,
|
|
"loss": 7.2444,
|
|
"mean_token_accuracy": 0.09039352908730507,
|
|
"num_tokens": 2567472.0,
|
|
"step": 1175
|
|
},
|
|
{
|
|
"entropy": 7.473226404190063,
|
|
"epoch": 1.2594767752269087,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004994911590988421,
|
|
"loss": 7.2288,
|
|
"mean_token_accuracy": 0.09728409722447395,
|
|
"num_tokens": 2578030.0,
|
|
"step": 1180
|
|
},
|
|
{
|
|
"entropy": 7.461479711532593,
|
|
"epoch": 1.2648158035237587,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004994623466943816,
|
|
"loss": 7.2471,
|
|
"mean_token_accuracy": 0.093011786788702,
|
|
"num_tokens": 2589925.0,
|
|
"step": 1185
|
|
},
|
|
{
|
|
"entropy": 7.40461311340332,
|
|
"epoch": 1.2701548318206086,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004994327418422922,
|
|
"loss": 7.1787,
|
|
"mean_token_accuracy": 0.09393652454018593,
|
|
"num_tokens": 2601825.0,
|
|
"step": 1190
|
|
},
|
|
{
|
|
"entropy": 7.403125238418579,
|
|
"epoch": 1.2754938601174586,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004994023446470912,
|
|
"loss": 7.1985,
|
|
"mean_token_accuracy": 0.09799938648939133,
|
|
"num_tokens": 2612726.0,
|
|
"step": 1195
|
|
},
|
|
{
|
|
"entropy": 7.348568964004516,
|
|
"epoch": 1.2808328884143085,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004993711552160936,
|
|
"loss": 7.1336,
|
|
"mean_token_accuracy": 0.09304829761385917,
|
|
"num_tokens": 2623753.0,
|
|
"step": 1200
|
|
},
|
|
{
|
|
"entropy": 7.404272222518921,
|
|
"epoch": 1.2861719167111585,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.0004993391736594117,
|
|
"loss": 7.1593,
|
|
"mean_token_accuracy": 0.0968549020588398,
|
|
"num_tokens": 2635275.0,
|
|
"step": 1205
|
|
},
|
|
{
|
|
"entropy": 7.480840063095092,
|
|
"epoch": 1.2915109450080084,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004993064000899538,
|
|
"loss": 7.2019,
|
|
"mean_token_accuracy": 0.09882493540644646,
|
|
"num_tokens": 2646329.0,
|
|
"step": 1210
|
|
},
|
|
{
|
|
"entropy": 7.275875473022461,
|
|
"epoch": 1.2968499733048584,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.0004992728346234246,
|
|
"loss": 7.2084,
|
|
"mean_token_accuracy": 0.09495408535003662,
|
|
"num_tokens": 2657629.0,
|
|
"step": 1215
|
|
},
|
|
{
|
|
"entropy": 7.453315734863281,
|
|
"epoch": 1.3021890016017084,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004992384773783244,
|
|
"loss": 7.1886,
|
|
"mean_token_accuracy": 0.09616189375519753,
|
|
"num_tokens": 2667746.0,
|
|
"step": 1220
|
|
},
|
|
{
|
|
"entropy": 7.27887864112854,
|
|
"epoch": 1.3075280298985585,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004992033284759491,
|
|
"loss": 7.1507,
|
|
"mean_token_accuracy": 0.10439181700348854,
|
|
"num_tokens": 2678157.0,
|
|
"step": 1225
|
|
},
|
|
{
|
|
"entropy": 7.484576034545898,
|
|
"epoch": 1.3128670581954085,
|
|
"grad_norm": 1.0703125,
|
|
"learning_rate": 0.0004991673880403891,
|
|
"loss": 7.2097,
|
|
"mean_token_accuracy": 0.09509778320789337,
|
|
"num_tokens": 2689319.0,
|
|
"step": 1230
|
|
},
|
|
{
|
|
"entropy": 7.382059717178345,
|
|
"epoch": 1.3182060864922585,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004991306561985294,
|
|
"loss": 7.1406,
|
|
"mean_token_accuracy": 0.09976356402039528,
|
|
"num_tokens": 2699718.0,
|
|
"step": 1235
|
|
},
|
|
{
|
|
"entropy": 7.359973287582397,
|
|
"epoch": 1.3235451147891084,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004990931330800492,
|
|
"loss": 7.19,
|
|
"mean_token_accuracy": 0.09778250828385353,
|
|
"num_tokens": 2710888.0,
|
|
"step": 1240
|
|
},
|
|
{
|
|
"entropy": 7.3708795547485355,
|
|
"epoch": 1.3288841430859584,
|
|
"grad_norm": 1.0078125,
|
|
"learning_rate": 0.0004990548188174211,
|
|
"loss": 7.1108,
|
|
"mean_token_accuracy": 0.0972976379096508,
|
|
"num_tokens": 2722024.0,
|
|
"step": 1245
|
|
},
|
|
{
|
|
"entropy": 7.322368574142456,
|
|
"epoch": 1.3342231713828083,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004990157135459105,
|
|
"loss": 7.0693,
|
|
"mean_token_accuracy": 0.09633990824222564,
|
|
"num_tokens": 2731780.0,
|
|
"step": 1250
|
|
},
|
|
{
|
|
"entropy": 7.395527935028076,
|
|
"epoch": 1.3395621996796583,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.0004989758174035758,
|
|
"loss": 7.1745,
|
|
"mean_token_accuracy": 0.0960126519203186,
|
|
"num_tokens": 2742524.0,
|
|
"step": 1255
|
|
},
|
|
{
|
|
"entropy": 7.409232425689697,
|
|
"epoch": 1.3449012279765082,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004989351305312676,
|
|
"loss": 7.1378,
|
|
"mean_token_accuracy": 0.09832295700907707,
|
|
"num_tokens": 2752077.0,
|
|
"step": 1260
|
|
},
|
|
{
|
|
"entropy": 7.322344446182251,
|
|
"epoch": 1.3502402562733582,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004988936530726276,
|
|
"loss": 7.1473,
|
|
"mean_token_accuracy": 0.09714090898633003,
|
|
"num_tokens": 2763625.0,
|
|
"step": 1265
|
|
},
|
|
{
|
|
"entropy": 7.293940925598145,
|
|
"epoch": 1.3555792845702082,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004988513851740889,
|
|
"loss": 7.1327,
|
|
"mean_token_accuracy": 0.10281604751944542,
|
|
"num_tokens": 2775262.0,
|
|
"step": 1270
|
|
},
|
|
{
|
|
"entropy": 7.3934331893920895,
|
|
"epoch": 1.3609183128670583,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004988083269848753,
|
|
"loss": 7.1964,
|
|
"mean_token_accuracy": 0.09679302722215652,
|
|
"num_tokens": 2786449.0,
|
|
"step": 1275
|
|
},
|
|
{
|
|
"entropy": 7.385905408859253,
|
|
"epoch": 1.3662573411639083,
|
|
"grad_norm": 0.80078125,
|
|
"learning_rate": 0.0004987644786570003,
|
|
"loss": 7.1696,
|
|
"mean_token_accuracy": 0.09244814142584801,
|
|
"num_tokens": 2796891.0,
|
|
"step": 1280
|
|
},
|
|
{
|
|
"entropy": 7.33904447555542,
|
|
"epoch": 1.3715963694607582,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004987198403452674,
|
|
"loss": 7.1607,
|
|
"mean_token_accuracy": 0.09559275880455971,
|
|
"num_tokens": 2806926.0,
|
|
"step": 1285
|
|
},
|
|
{
|
|
"entropy": 7.3863444328308105,
|
|
"epoch": 1.3769353977576082,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.0004986744122072688,
|
|
"loss": 7.0899,
|
|
"mean_token_accuracy": 0.09752057045698166,
|
|
"num_tokens": 2817750.0,
|
|
"step": 1290
|
|
},
|
|
{
|
|
"entropy": 7.180061292648316,
|
|
"epoch": 1.3822744260544582,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004986281944033851,
|
|
"loss": 7.1001,
|
|
"mean_token_accuracy": 0.10428606942296029,
|
|
"num_tokens": 2827770.0,
|
|
"step": 1295
|
|
},
|
|
{
|
|
"entropy": 7.401736879348755,
|
|
"epoch": 1.3876134543513081,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004985811870967848,
|
|
"loss": 7.1361,
|
|
"mean_token_accuracy": 0.10217376500368118,
|
|
"num_tokens": 2839583.0,
|
|
"step": 1300
|
|
},
|
|
{
|
|
"entropy": 7.363368988037109,
|
|
"epoch": 1.392952482648158,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004985333904534237,
|
|
"loss": 7.1358,
|
|
"mean_token_accuracy": 0.10174108073115348,
|
|
"num_tokens": 2850621.0,
|
|
"step": 1305
|
|
},
|
|
{
|
|
"entropy": 7.260199737548828,
|
|
"epoch": 1.398291510945008,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004984848046420445,
|
|
"loss": 7.1302,
|
|
"mean_token_accuracy": 0.09183874055743217,
|
|
"num_tokens": 2862287.0,
|
|
"step": 1310
|
|
},
|
|
{
|
|
"entropy": 7.392441749572754,
|
|
"epoch": 1.403630539241858,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.0004984354298341756,
|
|
"loss": 7.1278,
|
|
"mean_token_accuracy": 0.10441029965877532,
|
|
"num_tokens": 2871769.0,
|
|
"step": 1315
|
|
},
|
|
{
|
|
"entropy": 7.305068445205689,
|
|
"epoch": 1.408969567538708,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004983852662041312,
|
|
"loss": 7.1009,
|
|
"mean_token_accuracy": 0.0979885146021843,
|
|
"num_tokens": 2882590.0,
|
|
"step": 1320
|
|
},
|
|
{
|
|
"entropy": 7.323603248596191,
|
|
"epoch": 1.414308595835558,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.00049833431392901,
|
|
"loss": 7.1618,
|
|
"mean_token_accuracy": 0.09200560078024864,
|
|
"num_tokens": 2893903.0,
|
|
"step": 1325
|
|
},
|
|
{
|
|
"entropy": 7.257247972488403,
|
|
"epoch": 1.4196476241324079,
|
|
"grad_norm": 1.0859375,
|
|
"learning_rate": 0.0004982825731886956,
|
|
"loss": 7.0838,
|
|
"mean_token_accuracy": 0.09896045774221421,
|
|
"num_tokens": 2904532.0,
|
|
"step": 1330
|
|
},
|
|
{
|
|
"entropy": 7.321243238449097,
|
|
"epoch": 1.4249866524292578,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004982300441658546,
|
|
"loss": 7.0602,
|
|
"mean_token_accuracy": 0.09538981467485427,
|
|
"num_tokens": 2916340.0,
|
|
"step": 1335
|
|
},
|
|
{
|
|
"entropy": 7.252778673171997,
|
|
"epoch": 1.4303256807261078,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004981767270459367,
|
|
"loss": 7.0686,
|
|
"mean_token_accuracy": 0.10086248964071273,
|
|
"num_tokens": 2926424.0,
|
|
"step": 1340
|
|
},
|
|
{
|
|
"entropy": 7.384427738189697,
|
|
"epoch": 1.4356647090229577,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004981226220171744,
|
|
"loss": 7.1487,
|
|
"mean_token_accuracy": 0.09660428166389465,
|
|
"num_tokens": 2937620.0,
|
|
"step": 1345
|
|
},
|
|
{
|
|
"entropy": 7.306267547607422,
|
|
"epoch": 1.4410037373198077,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004980677292705811,
|
|
"loss": 7.0283,
|
|
"mean_token_accuracy": 0.10887585952877998,
|
|
"num_tokens": 2948274.0,
|
|
"step": 1350
|
|
},
|
|
{
|
|
"entropy": 7.251709699630737,
|
|
"epoch": 1.4463427656166576,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004980120489999519,
|
|
"loss": 7.1382,
|
|
"mean_token_accuracy": 0.09639003127813339,
|
|
"num_tokens": 2958986.0,
|
|
"step": 1355
|
|
},
|
|
{
|
|
"entropy": 7.4081559658050535,
|
|
"epoch": 1.4516817939135078,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004979555814018617,
|
|
"loss": 7.162,
|
|
"mean_token_accuracy": 0.09467695504426957,
|
|
"num_tokens": 2970933.0,
|
|
"step": 1360
|
|
},
|
|
{
|
|
"entropy": 7.283700609207154,
|
|
"epoch": 1.4570208222103578,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.000497898326675665,
|
|
"loss": 7.1182,
|
|
"mean_token_accuracy": 0.10104830637574196,
|
|
"num_tokens": 2981197.0,
|
|
"step": 1365
|
|
},
|
|
{
|
|
"entropy": 7.330152797698974,
|
|
"epoch": 1.4623598505072077,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004978402850234956,
|
|
"loss": 7.1262,
|
|
"mean_token_accuracy": 0.10599048286676407,
|
|
"num_tokens": 2993732.0,
|
|
"step": 1370
|
|
},
|
|
{
|
|
"entropy": 7.264380216598511,
|
|
"epoch": 1.4676988788040577,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004977814566502651,
|
|
"loss": 7.071,
|
|
"mean_token_accuracy": 0.10518286749720573,
|
|
"num_tokens": 3004617.0,
|
|
"step": 1375
|
|
},
|
|
{
|
|
"entropy": 7.271407747268677,
|
|
"epoch": 1.4730379071009077,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004977218417636629,
|
|
"loss": 7.0138,
|
|
"mean_token_accuracy": 0.10099183842539787,
|
|
"num_tokens": 3015058.0,
|
|
"step": 1380
|
|
},
|
|
{
|
|
"entropy": 7.180662345886231,
|
|
"epoch": 1.4783769353977576,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004976614405741546,
|
|
"loss": 7.0199,
|
|
"mean_token_accuracy": 0.10256859883666039,
|
|
"num_tokens": 3025995.0,
|
|
"step": 1385
|
|
},
|
|
{
|
|
"entropy": 7.355836534500122,
|
|
"epoch": 1.4837159636946076,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004976002532949823,
|
|
"loss": 7.0472,
|
|
"mean_token_accuracy": 0.1040610671043396,
|
|
"num_tokens": 3036060.0,
|
|
"step": 1390
|
|
},
|
|
{
|
|
"entropy": 7.186236763000489,
|
|
"epoch": 1.4890549919914575,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004975382801421631,
|
|
"loss": 7.0578,
|
|
"mean_token_accuracy": 0.1070075087249279,
|
|
"num_tokens": 3047012.0,
|
|
"step": 1395
|
|
},
|
|
{
|
|
"entropy": 7.2215704917907715,
|
|
"epoch": 1.4943940202883075,
|
|
"grad_norm": 0.953125,
|
|
"learning_rate": 0.0004974755213344886,
|
|
"loss": 6.9844,
|
|
"mean_token_accuracy": 0.1064968764781952,
|
|
"num_tokens": 3058186.0,
|
|
"step": 1400
|
|
},
|
|
{
|
|
"entropy": 7.222527360916137,
|
|
"epoch": 1.4997330485851574,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004974119770935242,
|
|
"loss": 6.9946,
|
|
"mean_token_accuracy": 0.10440263673663139,
|
|
"num_tokens": 3068566.0,
|
|
"step": 1405
|
|
},
|
|
{
|
|
"entropy": 7.2857832431793215,
|
|
"epoch": 1.5050720768820076,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.0004973476476436081,
|
|
"loss": 7.0199,
|
|
"mean_token_accuracy": 0.10458874180912972,
|
|
"num_tokens": 3079588.0,
|
|
"step": 1410
|
|
},
|
|
{
|
|
"entropy": 7.08985447883606,
|
|
"epoch": 1.5104111051788576,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004972825332118505,
|
|
"loss": 7.0023,
|
|
"mean_token_accuracy": 0.10629142746329308,
|
|
"num_tokens": 3092376.0,
|
|
"step": 1415
|
|
},
|
|
{
|
|
"entropy": 7.245244026184082,
|
|
"epoch": 1.5157501334757075,
|
|
"grad_norm": 0.8671875,
|
|
"learning_rate": 0.0004972166340281331,
|
|
"loss": 7.0461,
|
|
"mean_token_accuracy": 0.10235923826694489,
|
|
"num_tokens": 3103439.0,
|
|
"step": 1420
|
|
},
|
|
{
|
|
"entropy": 7.26457109451294,
|
|
"epoch": 1.5210891617725575,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004971499503251083,
|
|
"loss": 7.1008,
|
|
"mean_token_accuracy": 0.10314854681491852,
|
|
"num_tokens": 3115007.0,
|
|
"step": 1425
|
|
},
|
|
{
|
|
"entropy": 7.262210988998413,
|
|
"epoch": 1.5264281900694074,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004970824823381976,
|
|
"loss": 7.0539,
|
|
"mean_token_accuracy": 0.10062195062637329,
|
|
"num_tokens": 3125632.0,
|
|
"step": 1430
|
|
},
|
|
{
|
|
"entropy": 7.242954254150391,
|
|
"epoch": 1.5317672183662574,
|
|
"grad_norm": 0.98828125,
|
|
"learning_rate": 0.000497014230305592,
|
|
"loss": 7.0316,
|
|
"mean_token_accuracy": 0.10418535098433494,
|
|
"num_tokens": 3135825.0,
|
|
"step": 1435
|
|
},
|
|
{
|
|
"entropy": 7.2584309577941895,
|
|
"epoch": 1.5371062466631074,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004969451944682501,
|
|
"loss": 7.0369,
|
|
"mean_token_accuracy": 0.10161466002464295,
|
|
"num_tokens": 3147689.0,
|
|
"step": 1440
|
|
},
|
|
{
|
|
"entropy": 7.136695671081543,
|
|
"epoch": 1.5424452749599573,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004968753750698978,
|
|
"loss": 6.993,
|
|
"mean_token_accuracy": 0.10334522351622581,
|
|
"num_tokens": 3158754.0,
|
|
"step": 1445
|
|
},
|
|
{
|
|
"entropy": 7.298619794845581,
|
|
"epoch": 1.5477843032568073,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004968047723570273,
|
|
"loss": 7.0495,
|
|
"mean_token_accuracy": 0.09448943957686425,
|
|
"num_tokens": 3169332.0,
|
|
"step": 1450
|
|
},
|
|
{
|
|
"entropy": 7.218168115615844,
|
|
"epoch": 1.5531233315536572,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004967333865788961,
|
|
"loss": 7.124,
|
|
"mean_token_accuracy": 0.10236010178923607,
|
|
"num_tokens": 3180761.0,
|
|
"step": 1455
|
|
},
|
|
{
|
|
"entropy": 7.4003571510314945,
|
|
"epoch": 1.5584623598505072,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004966612179875267,
|
|
"loss": 7.0762,
|
|
"mean_token_accuracy": 0.09902083575725555,
|
|
"num_tokens": 3191635.0,
|
|
"step": 1460
|
|
},
|
|
{
|
|
"entropy": 7.129079866409302,
|
|
"epoch": 1.5638013881473571,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004965882668377048,
|
|
"loss": 7.0424,
|
|
"mean_token_accuracy": 0.10169995054602624,
|
|
"num_tokens": 3202452.0,
|
|
"step": 1465
|
|
},
|
|
{
|
|
"entropy": 7.34759840965271,
|
|
"epoch": 1.569140416444207,
|
|
"grad_norm": 1.078125,
|
|
"learning_rate": 0.0004965145333869789,
|
|
"loss": 6.971,
|
|
"mean_token_accuracy": 0.10686884224414825,
|
|
"num_tokens": 3213298.0,
|
|
"step": 1470
|
|
},
|
|
{
|
|
"entropy": 7.113559675216675,
|
|
"epoch": 1.574479444741057,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004964400178956595,
|
|
"loss": 7.0056,
|
|
"mean_token_accuracy": 0.10754135549068451,
|
|
"num_tokens": 3224314.0,
|
|
"step": 1475
|
|
},
|
|
{
|
|
"entropy": 7.2728907585144045,
|
|
"epoch": 1.579818473037907,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004963647206268181,
|
|
"loss": 7.0051,
|
|
"mean_token_accuracy": 0.1074408732354641,
|
|
"num_tokens": 3235660.0,
|
|
"step": 1480
|
|
},
|
|
{
|
|
"entropy": 7.222253370285034,
|
|
"epoch": 1.585157501334757,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.000496288641846286,
|
|
"loss": 7.0763,
|
|
"mean_token_accuracy": 0.09584841802716255,
|
|
"num_tokens": 3246856.0,
|
|
"step": 1485
|
|
},
|
|
{
|
|
"entropy": 7.304793119430542,
|
|
"epoch": 1.590496529631607,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.0004962117818226538,
|
|
"loss": 7.1116,
|
|
"mean_token_accuracy": 0.10176649838685989,
|
|
"num_tokens": 3258692.0,
|
|
"step": 1490
|
|
},
|
|
{
|
|
"entropy": 7.17184624671936,
|
|
"epoch": 1.5958355579284569,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.00049613414082727,
|
|
"loss": 7.0102,
|
|
"mean_token_accuracy": 0.10286101996898651,
|
|
"num_tokens": 3270577.0,
|
|
"step": 1495
|
|
},
|
|
{
|
|
"entropy": 7.200855493545532,
|
|
"epoch": 1.6011745862253068,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004960557191342405,
|
|
"loss": 6.9376,
|
|
"mean_token_accuracy": 0.10783654004335404,
|
|
"num_tokens": 3281262.0,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"epoch": 1.6011745862253068,
|
|
"eval_entropy": 7.068539691281009,
|
|
"eval_loss": 7.1312689781188965,
|
|
"eval_mean_token_accuracy": 0.10161492010796225,
|
|
"eval_num_tokens": 3281262.0,
|
|
"eval_runtime": 1.8156,
|
|
"eval_samples_per_second": 1696.444,
|
|
"eval_steps_per_second": 212.055,
|
|
"step": 1500
|
|
},
|
|
{
|
|
"entropy": 7.21118049621582,
|
|
"epoch": 1.606513614522157,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.000495976517020427,
|
|
"loss": 6.9955,
|
|
"mean_token_accuracy": 0.10954990237951279,
|
|
"num_tokens": 3291552.0,
|
|
"step": 1505
|
|
},
|
|
{
|
|
"entropy": 7.275783395767212,
|
|
"epoch": 1.611852642819007,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004958965347654469,
|
|
"loss": 7.0386,
|
|
"mean_token_accuracy": 0.10242254436016082,
|
|
"num_tokens": 3301977.0,
|
|
"step": 1510
|
|
},
|
|
{
|
|
"entropy": 7.121581172943115,
|
|
"epoch": 1.617191671115857,
|
|
"grad_norm": 0.8828125,
|
|
"learning_rate": 0.0004958157726516715,
|
|
"loss": 6.9699,
|
|
"mean_token_accuracy": 0.10966179370880128,
|
|
"num_tokens": 3312783.0,
|
|
"step": 1515
|
|
},
|
|
{
|
|
"entropy": 7.285579776763916,
|
|
"epoch": 1.622530699412707,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004957342309642256,
|
|
"loss": 6.9964,
|
|
"mean_token_accuracy": 0.10571648105978966,
|
|
"num_tokens": 3323895.0,
|
|
"step": 1520
|
|
},
|
|
{
|
|
"entropy": 7.104770946502685,
|
|
"epoch": 1.6278697277095568,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004956519099909857,
|
|
"loss": 6.9295,
|
|
"mean_token_accuracy": 0.10804956257343293,
|
|
"num_tokens": 3334103.0,
|
|
"step": 1525
|
|
},
|
|
{
|
|
"entropy": 7.208183097839355,
|
|
"epoch": 1.6332087560064068,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004955688100225803,
|
|
"loss": 6.9827,
|
|
"mean_token_accuracy": 0.10653971955180168,
|
|
"num_tokens": 3346279.0,
|
|
"step": 1530
|
|
},
|
|
{
|
|
"entropy": 7.092655944824219,
|
|
"epoch": 1.6385477843032568,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004954849313523875,
|
|
"loss": 6.9509,
|
|
"mean_token_accuracy": 0.10072449594736099,
|
|
"num_tokens": 3357582.0,
|
|
"step": 1535
|
|
},
|
|
{
|
|
"entropy": 7.243341445922852,
|
|
"epoch": 1.643886812600107,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004954002742765347,
|
|
"loss": 6.9531,
|
|
"mean_token_accuracy": 0.10738401338458062,
|
|
"num_tokens": 3369297.0,
|
|
"step": 1540
|
|
},
|
|
{
|
|
"entropy": 7.092299699783325,
|
|
"epoch": 1.649225840896957,
|
|
"grad_norm": 0.84765625,
|
|
"learning_rate": 0.0004953148390938973,
|
|
"loss": 7.0684,
|
|
"mean_token_accuracy": 0.10321348384022713,
|
|
"num_tokens": 3382430.0,
|
|
"step": 1545
|
|
},
|
|
{
|
|
"entropy": 7.257068061828614,
|
|
"epoch": 1.6545648691938069,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.000495228626106098,
|
|
"loss": 7.0312,
|
|
"mean_token_accuracy": 0.11251498833298683,
|
|
"num_tokens": 3393809.0,
|
|
"step": 1550
|
|
},
|
|
{
|
|
"entropy": 7.158068895339966,
|
|
"epoch": 1.6599038974906568,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004951416356175053,
|
|
"loss": 7.0077,
|
|
"mean_token_accuracy": 0.10660256892442703,
|
|
"num_tokens": 3405164.0,
|
|
"step": 1555
|
|
},
|
|
{
|
|
"entropy": 7.227698040008545,
|
|
"epoch": 1.6652429257875068,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004950538679352327,
|
|
"loss": 7.0212,
|
|
"mean_token_accuracy": 0.09924901276826859,
|
|
"num_tokens": 3416091.0,
|
|
"step": 1560
|
|
},
|
|
{
|
|
"entropy": 7.283638668060303,
|
|
"epoch": 1.6705819540843567,
|
|
"grad_norm": 0.83203125,
|
|
"learning_rate": 0.0004949653233691373,
|
|
"loss": 7.0712,
|
|
"mean_token_accuracy": 0.10288623869419097,
|
|
"num_tokens": 3427675.0,
|
|
"step": 1565
|
|
},
|
|
{
|
|
"entropy": 7.135433340072632,
|
|
"epoch": 1.6759209823812067,
|
|
"grad_norm": 0.97265625,
|
|
"learning_rate": 0.0004948760022318194,
|
|
"loss": 6.9453,
|
|
"mean_token_accuracy": 0.11070238053798676,
|
|
"num_tokens": 3438119.0,
|
|
"step": 1570
|
|
},
|
|
{
|
|
"entropy": 7.199069452285767,
|
|
"epoch": 1.6812600106780566,
|
|
"grad_norm": 0.91796875,
|
|
"learning_rate": 0.0004947859048386205,
|
|
"loss": 6.9806,
|
|
"mean_token_accuracy": 0.10209845751523972,
|
|
"num_tokens": 3449151.0,
|
|
"step": 1575
|
|
},
|
|
{
|
|
"entropy": 7.132083177566528,
|
|
"epoch": 1.6865990389749066,
|
|
"grad_norm": 0.9609375,
|
|
"learning_rate": 0.0004946950315076225,
|
|
"loss": 6.8718,
|
|
"mean_token_accuracy": 0.10959059745073318,
|
|
"num_tokens": 3459697.0,
|
|
"step": 1580
|
|
},
|
|
{
|
|
"entropy": 7.093738555908203,
|
|
"epoch": 1.6919380672717566,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004946033825596474,
|
|
"loss": 6.9422,
|
|
"mean_token_accuracy": 0.10823669135570527,
|
|
"num_tokens": 3470409.0,
|
|
"step": 1585
|
|
},
|
|
{
|
|
"entropy": 7.1705444812774655,
|
|
"epoch": 1.6972770955686065,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004945109583182548,
|
|
"loss": 6.9651,
|
|
"mean_token_accuracy": 0.1007260449230671,
|
|
"num_tokens": 3481515.0,
|
|
"step": 1590
|
|
},
|
|
{
|
|
"entropy": 7.167929172515869,
|
|
"epoch": 1.7026161238654565,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004944177591097415,
|
|
"loss": 6.9069,
|
|
"mean_token_accuracy": 0.10608908087015152,
|
|
"num_tokens": 3492683.0,
|
|
"step": 1595
|
|
},
|
|
{
|
|
"entropy": 7.120382785797119,
|
|
"epoch": 1.7079551521623064,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004943237852631404,
|
|
"loss": 6.9469,
|
|
"mean_token_accuracy": 0.10530265048146248,
|
|
"num_tokens": 3503067.0,
|
|
"step": 1600
|
|
},
|
|
{
|
|
"entropy": 7.19160099029541,
|
|
"epoch": 1.7132941804591564,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004942290371102193,
|
|
"loss": 6.9651,
|
|
"mean_token_accuracy": 0.10647035390138626,
|
|
"num_tokens": 3513611.0,
|
|
"step": 1605
|
|
},
|
|
{
|
|
"entropy": 7.181960487365723,
|
|
"epoch": 1.7186332087560063,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004941335149854794,
|
|
"loss": 6.9617,
|
|
"mean_token_accuracy": 0.10861775949597359,
|
|
"num_tokens": 3523802.0,
|
|
"step": 1610
|
|
},
|
|
{
|
|
"entropy": 7.190891218185425,
|
|
"epoch": 1.7239722370528563,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004940372192261546,
|
|
"loss": 6.9297,
|
|
"mean_token_accuracy": 0.10318318530917167,
|
|
"num_tokens": 3534764.0,
|
|
"step": 1615
|
|
},
|
|
{
|
|
"entropy": 7.121490812301635,
|
|
"epoch": 1.7293112653497063,
|
|
"grad_norm": 0.984375,
|
|
"learning_rate": 0.0004939401501722098,
|
|
"loss": 6.9932,
|
|
"mean_token_accuracy": 0.10756855309009553,
|
|
"num_tokens": 3544852.0,
|
|
"step": 1620
|
|
},
|
|
{
|
|
"entropy": 7.093603610992432,
|
|
"epoch": 1.7346502936465562,
|
|
"grad_norm": 0.875,
|
|
"learning_rate": 0.0004938423081663401,
|
|
"loss": 6.7933,
|
|
"mean_token_accuracy": 0.11004197001457214,
|
|
"num_tokens": 3555470.0,
|
|
"step": 1625
|
|
},
|
|
{
|
|
"entropy": 7.121168518066407,
|
|
"epoch": 1.7399893219434062,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.0004937436935539695,
|
|
"loss": 6.9935,
|
|
"mean_token_accuracy": 0.10067067667841911,
|
|
"num_tokens": 3567497.0,
|
|
"step": 1630
|
|
},
|
|
{
|
|
"entropy": 7.194992160797119,
|
|
"epoch": 1.7453283502402561,
|
|
"grad_norm": 0.85546875,
|
|
"learning_rate": 0.0004936443066832494,
|
|
"loss": 6.9725,
|
|
"mean_token_accuracy": 0.10623072907328605,
|
|
"num_tokens": 3578354.0,
|
|
"step": 1635
|
|
},
|
|
{
|
|
"entropy": 7.0331216812133786,
|
|
"epoch": 1.7506673785371063,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004935441479050579,
|
|
"loss": 6.8995,
|
|
"mean_token_accuracy": 0.11453714296221733,
|
|
"num_tokens": 3589369.0,
|
|
"step": 1640
|
|
},
|
|
{
|
|
"entropy": 7.2113524913787845,
|
|
"epoch": 1.7560064068339563,
|
|
"grad_norm": 0.828125,
|
|
"learning_rate": 0.0004934432175729981,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.10899419113993644,
|
|
"num_tokens": 3600396.0,
|
|
"step": 1645
|
|
},
|
|
{
|
|
"entropy": 7.116479682922363,
|
|
"epoch": 1.7613454351308062,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.000493341516043397,
|
|
"loss": 6.9588,
|
|
"mean_token_accuracy": 0.11054658442735672,
|
|
"num_tokens": 3610670.0,
|
|
"step": 1650
|
|
},
|
|
{
|
|
"entropy": 7.107414627075196,
|
|
"epoch": 1.7666844634276562,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004932390436753043,
|
|
"loss": 6.9206,
|
|
"mean_token_accuracy": 0.10563154965639114,
|
|
"num_tokens": 3620821.0,
|
|
"step": 1655
|
|
},
|
|
{
|
|
"entropy": 7.1294042587280275,
|
|
"epoch": 1.7720234917245061,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004931358008304911,
|
|
"loss": 6.8989,
|
|
"mean_token_accuracy": 0.11248534172773361,
|
|
"num_tokens": 3631469.0,
|
|
"step": 1660
|
|
},
|
|
{
|
|
"entropy": 7.078049898147583,
|
|
"epoch": 1.777362520021356,
|
|
"grad_norm": 0.9921875,
|
|
"learning_rate": 0.0004930317878734486,
|
|
"loss": 6.9114,
|
|
"mean_token_accuracy": 0.1081713691353798,
|
|
"num_tokens": 3641605.0,
|
|
"step": 1665
|
|
},
|
|
{
|
|
"entropy": 7.201143026351929,
|
|
"epoch": 1.782701548318206,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004929270051713866,
|
|
"loss": 6.9417,
|
|
"mean_token_accuracy": 0.1108034186065197,
|
|
"num_tokens": 3652107.0,
|
|
"step": 1670
|
|
},
|
|
{
|
|
"entropy": 7.0569343090057375,
|
|
"epoch": 1.7880405766150562,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004928214530942331,
|
|
"loss": 6.9729,
|
|
"mean_token_accuracy": 0.10586601719260216,
|
|
"num_tokens": 3663399.0,
|
|
"step": 1675
|
|
},
|
|
{
|
|
"entropy": 7.078392410278321,
|
|
"epoch": 1.7933796049119062,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004927151320146313,
|
|
"loss": 6.951,
|
|
"mean_token_accuracy": 0.10743931010365486,
|
|
"num_tokens": 3673638.0,
|
|
"step": 1680
|
|
},
|
|
{
|
|
"entropy": 7.162407350540161,
|
|
"epoch": 1.7987186332087561,
|
|
"grad_norm": 0.9296875,
|
|
"learning_rate": 0.0004926080423079404,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.10548929050564766,
|
|
"num_tokens": 3684459.0,
|
|
"step": 1685
|
|
},
|
|
{
|
|
"entropy": 7.04658784866333,
|
|
"epoch": 1.804057661505606,
|
|
"grad_norm": 0.84375,
|
|
"learning_rate": 0.0004925001843522323,
|
|
"loss": 6.9414,
|
|
"mean_token_accuracy": 0.10741329342126846,
|
|
"num_tokens": 3696113.0,
|
|
"step": 1690
|
|
},
|
|
{
|
|
"entropy": 7.169419765472412,
|
|
"epoch": 1.809396689802456,
|
|
"grad_norm": 0.83984375,
|
|
"learning_rate": 0.0004923915585282916,
|
|
"loss": 6.9062,
|
|
"mean_token_accuracy": 0.10803867951035499,
|
|
"num_tokens": 3707345.0,
|
|
"step": 1695
|
|
},
|
|
{
|
|
"entropy": 7.098963403701783,
|
|
"epoch": 1.814735718099306,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004922821652196136,
|
|
"loss": 6.744,
|
|
"mean_token_accuracy": 0.11550131812691689,
|
|
"num_tokens": 3717573.0,
|
|
"step": 1700
|
|
},
|
|
{
|
|
"entropy": 7.035619926452637,
|
|
"epoch": 1.820074746396156,
|
|
"grad_norm": 0.93359375,
|
|
"learning_rate": 0.0004921720048124034,
|
|
"loss": 6.9383,
|
|
"mean_token_accuracy": 0.11059200093150139,
|
|
"num_tokens": 3726658.0,
|
|
"step": 1705
|
|
},
|
|
{
|
|
"entropy": 7.158073139190674,
|
|
"epoch": 1.825413774693006,
|
|
"grad_norm": 0.87890625,
|
|
"learning_rate": 0.0004920610776955742,
|
|
"loss": 6.9495,
|
|
"mean_token_accuracy": 0.11338067129254341,
|
|
"num_tokens": 3737769.0,
|
|
"step": 1710
|
|
},
|
|
{
|
|
"entropy": 7.071769857406617,
|
|
"epoch": 1.8307528029898559,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004919493842607456,
|
|
"loss": 6.8737,
|
|
"mean_token_accuracy": 0.11217370927333832,
|
|
"num_tokens": 3748608.0,
|
|
"step": 1715
|
|
},
|
|
{
|
|
"entropy": 7.165196466445923,
|
|
"epoch": 1.8360918312867058,
|
|
"grad_norm": 0.9765625,
|
|
"learning_rate": 0.0004918369249022433,
|
|
"loss": 6.9148,
|
|
"mean_token_accuracy": 0.11021506115794182,
|
|
"num_tokens": 3759063.0,
|
|
"step": 1720
|
|
},
|
|
{
|
|
"entropy": 7.131874465942383,
|
|
"epoch": 1.8414308595835558,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004917237000170966,
|
|
"loss": 6.8688,
|
|
"mean_token_accuracy": 0.10861963406205177,
|
|
"num_tokens": 3769622.0,
|
|
"step": 1725
|
|
},
|
|
{
|
|
"entropy": 7.098778629302979,
|
|
"epoch": 1.8467698878804057,
|
|
"grad_norm": 0.96875,
|
|
"learning_rate": 0.0004916097100050375,
|
|
"loss": 6.8727,
|
|
"mean_token_accuracy": 0.11213916838169098,
|
|
"num_tokens": 3780587.0,
|
|
"step": 1730
|
|
},
|
|
{
|
|
"entropy": 7.07066969871521,
|
|
"epoch": 1.8521089161772557,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004914949552684992,
|
|
"loss": 6.9236,
|
|
"mean_token_accuracy": 0.11197843477129936,
|
|
"num_tokens": 3792386.0,
|
|
"step": 1735
|
|
},
|
|
{
|
|
"entropy": 7.145204401016235,
|
|
"epoch": 1.8574479444741057,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000491379436212615,
|
|
"loss": 6.9396,
|
|
"mean_token_accuracy": 0.11039775982499123,
|
|
"num_tokens": 3803820.0,
|
|
"step": 1740
|
|
},
|
|
{
|
|
"entropy": 7.06385440826416,
|
|
"epoch": 1.8627869727709556,
|
|
"grad_norm": 0.8515625,
|
|
"learning_rate": 0.000491263153245216,
|
|
"loss": 6.8982,
|
|
"mean_token_accuracy": 0.1090501606464386,
|
|
"num_tokens": 3815187.0,
|
|
"step": 1745
|
|
},
|
|
{
|
|
"entropy": 7.122003555297852,
|
|
"epoch": 1.8681260010678056,
|
|
"grad_norm": 0.86328125,
|
|
"learning_rate": 0.0004911461067768309,
|
|
"loss": 6.9121,
|
|
"mean_token_accuracy": 0.10848403200507165,
|
|
"num_tokens": 3826933.0,
|
|
"step": 1750
|
|
},
|
|
{
|
|
"entropy": 7.051048755645752,
|
|
"epoch": 1.8734650293646555,
|
|
"grad_norm": 0.921875,
|
|
"learning_rate": 0.0004910282972206834,
|
|
"loss": 6.7997,
|
|
"mean_token_accuracy": 0.11381994634866714,
|
|
"num_tokens": 3836934.0,
|
|
"step": 1755
|
|
},
|
|
{
|
|
"entropy": 7.05472617149353,
|
|
"epoch": 1.8788040576615055,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.0004909097249926913,
|
|
"loss": 6.9048,
|
|
"mean_token_accuracy": 0.10961363092064857,
|
|
"num_tokens": 3848662.0,
|
|
"step": 1760
|
|
},
|
|
{
|
|
"entropy": 7.01745228767395,
|
|
"epoch": 1.8841430859583554,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004907903905114652,
|
|
"loss": 6.7385,
|
|
"mean_token_accuracy": 0.1175423063337803,
|
|
"num_tokens": 3860052.0,
|
|
"step": 1765
|
|
},
|
|
{
|
|
"entropy": 7.047714710235596,
|
|
"epoch": 1.8894821142552054,
|
|
"grad_norm": 0.890625,
|
|
"learning_rate": 0.0004906702941983066,
|
|
"loss": 6.8871,
|
|
"mean_token_accuracy": 0.10923791378736496,
|
|
"num_tokens": 3871867.0,
|
|
"step": 1770
|
|
},
|
|
{
|
|
"entropy": 7.121882438659668,
|
|
"epoch": 1.8948211425520554,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004905494364772065,
|
|
"loss": 6.9453,
|
|
"mean_token_accuracy": 0.1086169384419918,
|
|
"num_tokens": 3883612.0,
|
|
"step": 1775
|
|
},
|
|
{
|
|
"entropy": 7.0241187572479244,
|
|
"epoch": 1.9001601708489055,
|
|
"grad_norm": 0.94140625,
|
|
"learning_rate": 0.0004904278177748441,
|
|
"loss": 6.868,
|
|
"mean_token_accuracy": 0.10942395329475403,
|
|
"num_tokens": 3895520.0,
|
|
"step": 1780
|
|
},
|
|
{
|
|
"entropy": 7.096190881729126,
|
|
"epoch": 1.9054991991457555,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004903054385205854,
|
|
"loss": 6.8573,
|
|
"mean_token_accuracy": 0.10636932700872422,
|
|
"num_tokens": 3905898.0,
|
|
"step": 1785
|
|
},
|
|
{
|
|
"entropy": 7.115457010269165,
|
|
"epoch": 1.9108382274426055,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004901822991464811,
|
|
"loss": 6.9076,
|
|
"mean_token_accuracy": 0.10664125084877014,
|
|
"num_tokens": 3916339.0,
|
|
"step": 1790
|
|
},
|
|
{
|
|
"entropy": 7.09182596206665,
|
|
"epoch": 1.9161772557394554,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004900584000872655,
|
|
"loss": 6.8633,
|
|
"mean_token_accuracy": 0.11126209571957588,
|
|
"num_tokens": 3927131.0,
|
|
"step": 1795
|
|
},
|
|
{
|
|
"entropy": 6.981594228744507,
|
|
"epoch": 1.9215162840363054,
|
|
"grad_norm": 0.96484375,
|
|
"learning_rate": 0.0004899337417803551,
|
|
"loss": 6.8705,
|
|
"mean_token_accuracy": 0.11410257518291474,
|
|
"num_tokens": 3938415.0,
|
|
"step": 1800
|
|
},
|
|
{
|
|
"entropy": 7.157159852981567,
|
|
"epoch": 1.9268553123331553,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.000489808324665847,
|
|
"loss": 6.8829,
|
|
"mean_token_accuracy": 0.11425913646817207,
|
|
"num_tokens": 3949700.0,
|
|
"step": 1805
|
|
},
|
|
{
|
|
"entropy": 7.0690977573394775,
|
|
"epoch": 1.9321943406300055,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004896821491865167,
|
|
"loss": 6.9192,
|
|
"mean_token_accuracy": 0.10939718559384345,
|
|
"num_tokens": 3961879.0,
|
|
"step": 1810
|
|
},
|
|
{
|
|
"entropy": 7.1183741092681885,
|
|
"epoch": 1.9375333689268555,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004895552157878174,
|
|
"loss": 6.8921,
|
|
"mean_token_accuracy": 0.1135055162012577,
|
|
"num_tokens": 3972903.0,
|
|
"step": 1815
|
|
},
|
|
{
|
|
"entropy": 7.02628173828125,
|
|
"epoch": 1.9428723972237054,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.0004894275249178782,
|
|
"loss": 6.8852,
|
|
"mean_token_accuracy": 0.11946139633655548,
|
|
"num_tokens": 3984432.0,
|
|
"step": 1820
|
|
},
|
|
{
|
|
"entropy": 7.089619779586792,
|
|
"epoch": 1.9482114255205554,
|
|
"grad_norm": 0.94921875,
|
|
"learning_rate": 0.0004892990770275021,
|
|
"loss": 6.9504,
|
|
"mean_token_accuracy": 0.10337705612182617,
|
|
"num_tokens": 3996998.0,
|
|
"step": 1825
|
|
},
|
|
{
|
|
"entropy": 7.125215005874634,
|
|
"epoch": 1.9535504538174053,
|
|
"grad_norm": 0.89453125,
|
|
"learning_rate": 0.0004891698725701648,
|
|
"loss": 6.8477,
|
|
"mean_token_accuracy": 0.11870014742016792,
|
|
"num_tokens": 4007614.0,
|
|
"step": 1830
|
|
},
|
|
{
|
|
"entropy": 7.0385946273803714,
|
|
"epoch": 1.9588894821142553,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.0004890399120020131,
|
|
"loss": 6.8262,
|
|
"mean_token_accuracy": 0.10775638073682785,
|
|
"num_tokens": 4017645.0,
|
|
"step": 1835
|
|
},
|
|
{
|
|
"entropy": 7.1667829036712645,
|
|
"epoch": 1.9642285104111052,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004889091957818631,
|
|
"loss": 6.9244,
|
|
"mean_token_accuracy": 0.10332238003611564,
|
|
"num_tokens": 4029353.0,
|
|
"step": 1840
|
|
},
|
|
{
|
|
"entropy": 7.053739356994629,
|
|
"epoch": 1.9695675387079552,
|
|
"grad_norm": 0.87109375,
|
|
"learning_rate": 0.0004887777243711988,
|
|
"loss": 6.8718,
|
|
"mean_token_accuracy": 0.11049497202038765,
|
|
"num_tokens": 4040421.0,
|
|
"step": 1845
|
|
},
|
|
{
|
|
"entropy": 7.0194093704223635,
|
|
"epoch": 1.9749065670048052,
|
|
"grad_norm": 1.015625,
|
|
"learning_rate": 0.00048864549823417,
|
|
"loss": 6.8325,
|
|
"mean_token_accuracy": 0.1106531098484993,
|
|
"num_tokens": 4051858.0,
|
|
"step": 1850
|
|
},
|
|
{
|
|
"entropy": 7.106148815155029,
|
|
"epoch": 1.9802455953016551,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004885125178375915,
|
|
"loss": 6.8803,
|
|
"mean_token_accuracy": 0.1077659197151661,
|
|
"num_tokens": 4062063.0,
|
|
"step": 1855
|
|
},
|
|
{
|
|
"entropy": 7.0133593559265135,
|
|
"epoch": 1.985584623598505,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004883787836509405,
|
|
"loss": 6.6985,
|
|
"mean_token_accuracy": 0.11615452989935875,
|
|
"num_tokens": 4072479.0,
|
|
"step": 1860
|
|
},
|
|
{
|
|
"entropy": 7.0427937507629395,
|
|
"epoch": 1.990923651895355,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004882442961463558,
|
|
"loss": 6.8677,
|
|
"mean_token_accuracy": 0.11139330714941025,
|
|
"num_tokens": 4082912.0,
|
|
"step": 1865
|
|
},
|
|
{
|
|
"entropy": 7.077240419387818,
|
|
"epoch": 1.996262680192205,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.0004881090557986353,
|
|
"loss": 6.8434,
|
|
"mean_token_accuracy": 0.11269779428839684,
|
|
"num_tokens": 4093644.0,
|
|
"step": 1870
|
|
},
|
|
{
|
|
"entropy": 6.915128972795275,
|
|
"epoch": 2.00106780565937,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004879730630852353,
|
|
"loss": 6.7595,
|
|
"mean_token_accuracy": 0.11212469720178181,
|
|
"num_tokens": 4104114.0,
|
|
"step": 1875
|
|
},
|
|
{
|
|
"entropy": 7.109417867660523,
|
|
"epoch": 2.0064068339562198,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.00048783631848626764,
|
|
"loss": 6.6362,
|
|
"mean_token_accuracy": 0.1157489076256752,
|
|
"num_tokens": 4114847.0,
|
|
"step": 1880
|
|
},
|
|
{
|
|
"entropy": 6.914434337615967,
|
|
"epoch": 2.0117458622530697,
|
|
"grad_norm": 1.0234375,
|
|
"learning_rate": 0.0004876988224844992,
|
|
"loss": 6.4861,
|
|
"mean_token_accuracy": 0.12341470643877983,
|
|
"num_tokens": 4126078.0,
|
|
"step": 1885
|
|
},
|
|
{
|
|
"entropy": 7.015130233764649,
|
|
"epoch": 2.01708489054992,
|
|
"grad_norm": 0.88671875,
|
|
"learning_rate": 0.00048756057556534934,
|
|
"loss": 6.5112,
|
|
"mean_token_accuracy": 0.12700850293040275,
|
|
"num_tokens": 4136678.0,
|
|
"step": 1890
|
|
},
|
|
{
|
|
"entropy": 6.966462326049805,
|
|
"epoch": 2.02242391884677,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004874215782168884,
|
|
"loss": 6.5355,
|
|
"mean_token_accuracy": 0.1265234611928463,
|
|
"num_tokens": 4146794.0,
|
|
"step": 1895
|
|
},
|
|
{
|
|
"entropy": 7.0654607772827145,
|
|
"epoch": 2.02776294714362,
|
|
"grad_norm": 0.8984375,
|
|
"learning_rate": 0.0004872818309298365,
|
|
"loss": 6.571,
|
|
"mean_token_accuracy": 0.12405075505375862,
|
|
"num_tokens": 4156825.0,
|
|
"step": 1900
|
|
},
|
|
{
|
|
"entropy": 6.920158243179321,
|
|
"epoch": 2.03310197544047,
|
|
"grad_norm": 0.9140625,
|
|
"learning_rate": 0.00048714133419756065,
|
|
"loss": 6.5027,
|
|
"mean_token_accuracy": 0.11929825246334076,
|
|
"num_tokens": 4166765.0,
|
|
"step": 1905
|
|
},
|
|
{
|
|
"entropy": 7.046618700027466,
|
|
"epoch": 2.03844100373732,
|
|
"grad_norm": 0.98046875,
|
|
"learning_rate": 0.00048700008851607437,
|
|
"loss": 6.6338,
|
|
"mean_token_accuracy": 0.12172048911452293,
|
|
"num_tokens": 4177404.0,
|
|
"step": 1910
|
|
},
|
|
{
|
|
"entropy": 7.052170658111573,
|
|
"epoch": 2.04378003203417,
|
|
"grad_norm": 0.78125,
|
|
"learning_rate": 0.00048685809438403494,
|
|
"loss": 6.5971,
|
|
"mean_token_accuracy": 0.12037592306733132,
|
|
"num_tokens": 4188804.0,
|
|
"step": 1915
|
|
},
|
|
{
|
|
"entropy": 6.9027196884155275,
|
|
"epoch": 2.04911906033102,
|
|
"grad_norm": 0.9375,
|
|
"learning_rate": 0.0004867153523027422,
|
|
"loss": 6.4989,
|
|
"mean_token_accuracy": 0.1245061419904232,
|
|
"num_tokens": 4200009.0,
|
|
"step": 1920
|
|
},
|
|
{
|
|
"entropy": 7.009029340744019,
|
|
"epoch": 2.05445808862787,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004865718627761363,
|
|
"loss": 6.5549,
|
|
"mean_token_accuracy": 0.12319564968347549,
|
|
"num_tokens": 4210776.0,
|
|
"step": 1925
|
|
},
|
|
{
|
|
"entropy": 6.948365020751953,
|
|
"epoch": 2.05979711692472,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048642762631079646,
|
|
"loss": 6.5094,
|
|
"mean_token_accuracy": 0.13088932484388352,
|
|
"num_tokens": 4221412.0,
|
|
"step": 1930
|
|
},
|
|
{
|
|
"entropy": 6.867213582992553,
|
|
"epoch": 2.0651361452215697,
|
|
"grad_norm": 0.90234375,
|
|
"learning_rate": 0.0004862826434159388,
|
|
"loss": 6.5012,
|
|
"mean_token_accuracy": 0.12945465818047525,
|
|
"num_tokens": 4231747.0,
|
|
"step": 1935
|
|
},
|
|
{
|
|
"entropy": 6.948292064666748,
|
|
"epoch": 2.0704751735184197,
|
|
"grad_norm": 1.0,
|
|
"learning_rate": 0.00048613691460341456,
|
|
"loss": 6.5181,
|
|
"mean_token_accuracy": 0.12791630178689956,
|
|
"num_tokens": 4241942.0,
|
|
"step": 1940
|
|
},
|
|
{
|
|
"entropy": 6.955828285217285,
|
|
"epoch": 2.0758142018152697,
|
|
"grad_norm": 0.82421875,
|
|
"learning_rate": 0.00048599044038770856,
|
|
"loss": 6.6039,
|
|
"mean_token_accuracy": 0.12196211889386177,
|
|
"num_tokens": 4253151.0,
|
|
"step": 1945
|
|
},
|
|
{
|
|
"entropy": 6.93486852645874,
|
|
"epoch": 2.0811532301121196,
|
|
"grad_norm": 0.9453125,
|
|
"learning_rate": 0.0004858432212859372,
|
|
"loss": 6.5329,
|
|
"mean_token_accuracy": 0.12614453956484795,
|
|
"num_tokens": 4263579.0,
|
|
"step": 1950
|
|
},
|
|
{
|
|
"entropy": 6.9105040550231935,
|
|
"epoch": 2.0864922584089696,
|
|
"grad_norm": 1.0390625,
|
|
"learning_rate": 0.0004856952578178464,
|
|
"loss": 6.5513,
|
|
"mean_token_accuracy": 0.12219369858503341,
|
|
"num_tokens": 4274764.0,
|
|
"step": 1955
|
|
},
|
|
{
|
|
"entropy": 7.026956415176391,
|
|
"epoch": 2.0918312867058195,
|
|
"grad_norm": 0.95703125,
|
|
"learning_rate": 0.0004855465505058102,
|
|
"loss": 6.6513,
|
|
"mean_token_accuracy": 0.11608931571245193,
|
|
"num_tokens": 4286561.0,
|
|
"step": 1960
|
|
},
|
|
{
|
|
"entropy": 6.919632291793823,
|
|
"epoch": 2.0971703150026695,
|
|
"grad_norm": 1.1015625,
|
|
"learning_rate": 0.0004853970998748289,
|
|
"loss": 6.603,
|
|
"mean_token_accuracy": 0.1245616041123867,
|
|
"num_tokens": 4296101.0,
|
|
"step": 1965
|
|
},
|
|
{
|
|
"entropy": 6.937914419174194,
|
|
"epoch": 2.1025093432995194,
|
|
"grad_norm": 1.03125,
|
|
"learning_rate": 0.00048524690645252675,
|
|
"loss": 6.5519,
|
|
"mean_token_accuracy": 0.11856919229030609,
|
|
"num_tokens": 4307613.0,
|
|
"step": 1970
|
|
},
|
|
{
|
|
"entropy": 6.950258350372314,
|
|
"epoch": 2.1078483715963694,
|
|
"grad_norm": 0.92578125,
|
|
"learning_rate": 0.0004850959707691503,
|
|
"loss": 6.597,
|
|
"mean_token_accuracy": 0.12681118920445442,
|
|
"num_tokens": 4318042.0,
|
|
"step": 1975
|
|
},
|
|
{
|
|
"entropy": 6.949284172058105,
|
|
"epoch": 2.1131873998932194,
|
|
"grad_norm": 0.91015625,
|
|
"learning_rate": 0.000484944293357567,
|
|
"loss": 6.4691,
|
|
"mean_token_accuracy": 0.12812729254364968,
|
|
"num_tokens": 4328690.0,
|
|
"step": 1980
|
|
},
|
|
{
|
|
"entropy": 6.885101842880249,
|
|
"epoch": 2.1185264281900693,
|
|
"grad_norm": 1.0625,
|
|
"learning_rate": 0.00048479187475326254,
|
|
"loss": 6.5129,
|
|
"mean_token_accuracy": 0.1259954057633877,
|
|
"num_tokens": 4339188.0,
|
|
"step": 1985
|
|
},
|
|
{
|
|
"entropy": 6.827868366241455,
|
|
"epoch": 2.1238654564869193,
|
|
"grad_norm": 0.90625,
|
|
"learning_rate": 0.00048463871549433947,
|
|
"loss": 6.4728,
|
|
"mean_token_accuracy": 0.1304394595324993,
|
|
"num_tokens": 4350242.0,
|
|
"step": 1990
|
|
},
|
|
{
|
|
"entropy": 6.834375524520874,
|
|
"epoch": 2.1292044847837692,
|
|
"grad_norm": 0.8359375,
|
|
"learning_rate": 0.00048448481612151524,
|
|
"loss": 6.466,
|
|
"mean_token_accuracy": 0.13104819133877754,
|
|
"num_tokens": 4361138.0,
|
|
"step": 1995
|
|
},
|
|
{
|
|
"entropy": 6.958746910095215,
|
|
"epoch": 2.134543513080619,
|
|
"grad_norm": 0.99609375,
|
|
"learning_rate": 0.00048433017717812015,
|
|
"loss": 6.5523,
|
|
"mean_token_accuracy": 0.1202632024884224,
|
|
"num_tokens": 4372157.0,
|
|
"step": 2000
|
|
},
|
|
{
|
|
"epoch": 2.134543513080619,
|
|
"eval_entropy": 6.773753176107035,
|
|
"eval_loss": 6.934739112854004,
|
|
"eval_mean_token_accuracy": 0.11093968841936681,
|
|
"eval_num_tokens": 4372157.0,
|
|
"eval_runtime": 1.8151,
|
|
"eval_samples_per_second": 1696.873,
|
|
"eval_steps_per_second": 212.109,
|
|
"step": 2000
|
|
}
|
|
],
|
|
"logging_steps": 5,
|
|
"max_steps": 9360,
|
|
"num_input_tokens_seen": 0,
|
|
"num_train_epochs": 10,
|
|
"save_steps": 500,
|
|
"stateful_callbacks": {
|
|
"TrainerControl": {
|
|
"args": {
|
|
"should_epoch_stop": false,
|
|
"should_evaluate": false,
|
|
"should_log": false,
|
|
"should_save": true,
|
|
"should_training_stop": false
|
|
},
|
|
"attributes": {}
|
|
}
|
|
},
|
|
"total_flos": 948753236920320.0,
|
|
"train_batch_size": 16,
|
|
"trial_name": null,
|
|
"trial_params": null
|
|
}
|